Privileged by Design: AI Agents and the New Identity Risk to Production Systems – Shashwat Sehgal – RSAC26 #1

This episode is sponsored by
Full Segment Notes
Key Moments
  • 0:00 - RSAC 2026 Interview – AI Security Challenges with P0 Security
  • 0:50 - The Explosion of AI Tools & Agentic Frameworks
  • 01:46 - Why Security Teams Can’t Keep Up with AI Innovation
  • 02:21 - Identity Security Fundamentals: Connectivity, Auth & Authorization
  • 04:24 - Why Authorization is the Biggest Problem in AI Security
  • 04:49 - AI Agents vs Human Identity – Who Gets Access?
  • 05:59 - Securing AI Agents Before Production Deployment
  • 06:26 - Guardrails, Ownership & Governance for AI Systems
  • 08:46 - Just-In-Time Access & Short-Lived Credentials Explained
  • 09:31 - Shadow AI, Open Source Agents & Governance Gaps
  • 10:56 - Lessons from Cloud Security Failures (S3, IAM Risks)
  • 11:24 - Why Traditional Identity Tools Are Failing Today
  • 12:37 - API-Based Security: The New Identity Control Layer
  • 13:24 - Identity as the New Perimeter in Cybersecurity
  • 14:37 - Organizational Challenges in Identity & AI Security
Guest
CEO and co-founder at P0 Security

Shashwat Sehgal is the Co-Founder and CEO of P0 Security. He’s spent most of his career building security and observability products for developers, DevOps, and security teams.

Shashwat is passionate about solving the problem of cloud access security and helping security engineers’ control ‘who has access to sensitive resources in their clouds.’

He enjoys playing tennis, spending time with his family, teaching his son how to play chess, and geeking out on all things security.

Stay in the Know, No Smoke and Mirrors – Join Our Newsletter

You can skip this ad in 5 seconds