Your Logs Can't Tell You What an AI Agent Intended | Ashish Rajan
Security Unfiltered PodcastOctober 13, 202600:10:23

Your Logs Can't Tell You What an AI Agent Intended | Ashish Rajan

How do you secure millions of AI agents when your logs can't tell "is Joe an employee?" apart from "can I have Joe's password?" Joe South and Ashish Rajan (author of AI Security Engineering, host of the Cloud Security Podcast) dig into the agent security capabilities that actually work.

Joe walks through a control that interrogates an agent when it pulls 10 records after you asked for one, and Ashish explains why behavior and intent will matter more than log review, why an LLM firewall alone isn't enough, and why no vendor is going to build the internal context layer your security team needs across Slack, Teams, Confluence and SharePoint.

Full episode: https://www.youtube.com/watch?v=hLDD4FxTSRc
Ashish Rajan on LinkedIn: https://www.linkedin.com/in/ashishrajan/
AI Security Engineering (Wiley): https://www.amazon.com/dp/1394387687

Subscribe: https://www.youtube.com/@securityunfilteredpodcast
Show site: https://www.securityunfiltered.com
Ashish Rajan, security unfiltered, joe south, AI agent security, ai agents,
Powered by Beamly