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 Us
As AI becomes increasingly integrated into cloud security workflows, selecting the right combination of AI models and AI harnesses such as Claude Code, Codex, Cursor, and similar tools is critical. In this webcast, we’ll share the results of our comparative analysis of leading AI harnesses and foundation models across a range of cloud and security use cases.
Our goal was to determine which combinations perform best for different tasks by evaluating accuracy, speed, cost, reliability, and overall efficiency. We’ll discuss our methodology, the metrics we used, and the key trade offs we observed. Whether you’re building AI assisted security workflows, automating cloud operations, or evaluating AI platforms for enterprise use, this session will provide practical insights to help you choose the right solution for your needs.
Who should attend?
This session supports concepts from SEC545: GenAI and LLM Application Security. To learn more, browse upcoming sessions, and access your free course preview, Click Here.


Ahmed AbuGharbia, SANS Instructor and SEC545 author, helps practitioners secure generative AI systems by identifying risks, understanding model behavior, and applying practical security controls.
Read more about Ahmed AbuGharbia