SEC536: Adversarial AI - Penetration Testing AI Systems


AI models have shown they can break out of controlled environments and act entirely on their own, without a human in the loop. During the Hugging Face incident, that's exactly what happened, and the very AI tools built to help investigate ended up standing in the way. Our panel walks through how it unfolded and what it means for anyone whose IR plan assumes their tools will cooperate.
Watch the livestream to see exactly what happened, what changed, and what to do before it happens to you. After you watch, go deeper with the post-mortem brief SANS co-authored, published by Cloud Security Alliance with contributors from across the industry. Then take the two-minute self-assessment to see if your team is ready to handle an AI-run attack.
SANS AI security courses map to six GIAC certifications, from GIAC AI Platform Security (GAIPS) for engineers securing GenAI and LLM applications through GIAC AI Security Automation Engineer (GASAE) for practitioners automating red, blue, and purple team operations. Each course pairs hands-on labs in real AI environments with the certification exam, so the credential reflects skills practiced, not just concepts read about. Because AI security cuts across other disciplines, every certification below already has its own home under another focus area — this page exists for people specifically searching for AI security training, not as a duplicate.
Proves you can secure generative AI and LLM-powered applications against prompt injection, data leakage, and model abuse. For AI engineers, application security teams, and cloud architects.
Proves you can apply AI and automation across red, blue, and purple team operations. For security engineers and automation-focused practitioners.
Proves you can identify and exploit weaknesses in AI-enabled systems. For red teamers and AI security researchers.
Proves you can apply data science and machine learning to real security problems. For security engineers moving into AI/ML.
Proves you can build practical security automation tools with Python. For security engineers and analysts who want to code.
Proves you can test AI systems for adversarial weaknesses. For penetration testers working with AI-enabled systems.