session Replay

The Agent Problem: Where Access Meets Autonomy with Ashish Rajan

Join Ashish Rajan, host of the AI Security Podcast, and Varonis experts as they unpack the risks of AI agents that can access data, use tools, and act with minimal oversight. Discover real-world examples of agent failures, emerging attack paths, and the lessons security teams can apply before these risks turn into incidents.

wistia video thumbnail

What we cover:

  • Why agentic AI creates a new class of identity, data, and permission risk
  • What real-world incidents like HuggingFace and Claude sandbox escape reveal about agent behavior
  • How Ashish Rajan frames the next chapter of AI security for cloud and security teams
  • Where defenders should focus before deploying AI agents at scale
  • How to think about response when autonomous systems act outside their lane

Presenters:

Thumbnail-Headshot_BrianVecci_Square_Webinars

Brian Vecci

Field CTO, Varonis

Ashish Rajan new

Ashish Rajan

CISO and Host of the Cloud Security Podcast & AI Security Podcast

headshot-tal-peleg

Tal Peleg

DA Cloud Research Team Leader, Varonis

In this session, you'll learn:

  • The challenges with legacy DLP solutions
  • The threat actor’s advantage and common attack strategies for exfiltrating sensitive data
  • Actionable tips to improve your data security posture
  • Best practices to prevent data loss in the cloud and on-prem
  • Outcomes you can expect from adopting a next-gen DLP strategy
March 26 2025 - DFF - presenters