Autonomous AI agents like OpenClaw are quickly moving from experimentation into real enterprise environments. These systems are not just generating content. They are taking action, interacting with systems and operating with access to sensitive data and credentials.

This shift introduces a new class of machine identities with privileged access. Often, these identities are ungoverned, lack visibility and fall outside traditional security controls. As adoption accelerates, so does the risk.

In this session, CyberArk, a Palo Alto Networks company, brings frontline offensive security research to show how agentic AI is reshaping the identity attack surface. Andy Thompson from CyberArk Labs shares how modern adversaries exploit identity across AI, cloud and automated environments.

• Understand how AI agents function as machine identities with privileged access
• See where new identity risks emerge across secrets, APIs and cloud access
• Learn how attackers exploit AI-driven environments in real scenarios
• Explore practical ways to reduce risk without slowing innovation

Machine identities already outnumber humans by more than 80 to 1. As AI agents scale, security teams must rethink how they define privilege, enforce control and maintain visibility across every identity.

Presenters:

Andy Thompson

Senior Offensive Cybersecurity Research Evangelist at CyberArk Labs

REGISTER NOW