AI - SEC411: AI Security Principles & Practices: GenAI & LLM Defense

Master AI security from fundamentals through advanced defense in this hands-on course. Exploit tokenization vulnerabilities, execute prompt attacks, compromise RAG systems, and implement production-ready defenses - all in Docker-based labs with an integrated Learning Assistant.

Build essential GenAI and LLM security skills without prior AI experience. This course covers everything from tokenization and attack surface analysis to the OWASP Top 10 for LLMs.

Identify, exploit, and defend against AI-specific threats including prompt injection, jailbreaking, and RAG manipulation through progressive hands-on labs:

Lab 1: Tokenizer Security Explorer

• Level 1: Analyze basic tokenization and its security impact
• Level 2: Compare tokenization behaviors across multiple models
• Level 3: Execute tokenization attack techniques with guided examples
• Level 4: Deploy token-aware defenses against attack techniques
• Level 5: Complete comprehensive tokenization attack/defend scenarios

Lab 2: Prompt Injection Attack Lab

• Level 1: Master direct prompt injection fundamentals and detection techniques
• Level 2: Execute social engineering attacks using role manipulation
• Level 3: Apply obfuscation methods including Base64, ROT13, and Unicode
• Level 4: Perform system prompt extraction and context hijacking
• Level 5: Chain advanced jailbreaking and multi-turn attacks

Implement practical defenses for training pipelines, inference environments, and RAG systems across enterprise settings. Secure advanced AI architectures including agentic systems and reasoning models.

Apply frameworks such as MITRE ATLAS, OWASP Top 10 for LLMs, and NIST AI RMF in real-world security operations. Integrate AI security into existing SOC workflows and incident response processes.

Seth, SANS Faculty Fellow and author of SEC411, LDR414, and SEC511, combines cutting-edge consulting and education to equip defenders worldwide. Founder of Context Security and GSE #28, he brings clarity, humor, and purpose to cybersecurity training.

Course outcomes:

• Rapidly upskill cybersecurity teams on GenAI and LLM security
• Access continuous learning with new labs and modules during your 4-month access window
• Gain deployable skills through Docker-based hands-on labs
• Meet compliance requirements with NIST AI RMF, EU AI Act, and other standards