Are Your AI Agents Secure? Defending the Privileged Agent

Are the AI agents you’re developing truly secure?

AI agents that execute actions autonomously offer unprecedented value. But what about the “privileges“ granted to them to act “on behalf of the user“?

Improper privilege management for agents is no longer a theoretical problem—it’s a clear and present danger. An exploited AI agent with excessive privileges can lead to significant financial losses and devastating data breaches.

This session dives deep into the biggest pitfall in AI agent development: privilege and authorization. I will demystify the latest risks, such as Excessive Agency and Identity Abuse, and discuss defensive measures you can take to protect your AI agents from malicious actors. This is the critical security state that every development organization must understand before deploying AI agents into production.

Stop Rebuilding Authorisation: Accelerating AI Agent Development

with Kevin Akermanis

The demand for AI agents that take real action is immense. However, developers keep hitting a frustrating bottleneck: building the complex authorisation logic required to let agents act on behalf of users. Rebuilding custom authorisation for every new agent capability creates endless boilerplate, technical debt, and a massive integration mess. Instead of focusing on innovative AI features, engineering teams waste cycles reinventing access control and approval workflows. This session is all about maximising developer efficiency and eliminating rework. Discover how leveraging Auth0 for AI Agents handles the heavy lifting of dynamic permissions and seamless Human-in-the-Loop controls out of the box. We will explore how to stop building auth from scratch so you can ship high-impact AI agents with significantly less friction.