Thumbnail for The Post-Prompt World: Securing AI Agents That Think for Themselves

AI agents are increasingly trusted to act autonomously—but can be manipulated to violate security intent without exploiting code, models, or permissions. This RSAC session explores real-world attacks on agent reasoning that traditional controls miss, and presents a practical 90-day framework to

Watch the Full Video

Get access to videos on Machine Learning & Artificial Intelligence, exclusive research, RSA Conference sessions, webcasts, podcasts and more.

Create a Free Account

Already have an account?