SEC536: Adversarial AI - Penetration Testing AI Systems


The GIAC AI Penetration Tester (GAIPT) certification validates a practitioner's hands-on AI penetration testing skills for assessing and exploiting vulnerabilities in enterprise AI systems, including large language models, RAG pipelines, machine learning models, computer vision, and agentic AI.
GAIPT uses GIAC's CyberLive format, hands-on lab challenges rather than traditional multiple choice. Attempts are web-based and proctored, either remotely through ProctorU or onsite through PearsonVUE, and you have 120 days from activation to complete your attempt. See the GIAC GAIPT page for current exam details.
GIAC certifications are renewed on a recurring cycle through continuing education credits and a maintenance fee. For the current renewal requirements, see GIAC’s renewal page.
GAIPT fits penetration testers and red teamers, application security engineers and architects, data scientists and AI and ML platform engineers, security practitioners and cyber defenders who need to understand how AI systems work and can be exploited, and technical leaders who want more insight into the risks and mechanics of AI systems.
Taking SANS SEC536 maps directly to the certification objectives in GAIPT. SEC536: Adversarial AI - Penetration Testing AI Systems includes 15 hands-on labs run against a realistic, production-shaped environment, and the full syllabus is on the SEC536 course page.