SEC536: Adversarial AI - Penetration Testing AI Systems


Experience SANS training through course previews.
Learn MoreLet us help.
Contact usBecome a member for instant access to our free resources.
Sign UpWe're here to help.
Contact UsAttack your organization's AI before an adversary does by building prompt-injection, jailbreak, and adversarial-ML attack chains against LLM apps, RAG systems, and agents.
Construct prompt-injection and jailbreak sequences against LLM apps, RAG systems, and agents, not just single-shot prompts.
Execute evasion, poisoning, extraction, and inference attacks, and test agent tool permissions for confused-deputy and escalation paths.
Automate attacks in Python with frameworks like PyRIT, Garak, or Promptfoo, and write reproducible reports engineering teams can close.
Explore the courses below that are aligned with this job role.
Intermediate courses are designed for cybersecurity professionals with practical, hands-on experience.



There are numerous different roles in cybersecurity and where you fit depends on your interest level. SANS New to Cyber offers courses, certifications, and free resources for anyone interested in getting started in cybersecurity.