Posted on
in Presentations
AI agent security is not just about guardrails. This talk will introduce a framework for isolating LLM-specific vulnerabilities in agents. Based on 194k adversarial attacks across real agent scenarios and 30+ models, the talk will reveal surprising results: reasoning improves security, model size does not, and LLM choice meaningfully shapes agentic risk.
Access This and Other RSAC™ Conference Presentations with Your Free RSAC Membership
Your RSAC™ Membership also includes AI-powered summaries, mind maps, and slides for Conference presentations, Group Discussions with experts, and more.
Watch Now >>
Share With Your Community