Enterprise AI Security and Governance Summit 2026

As AI agents move from answering questions to taking action, they’re creating a new security challenge. Agents can access sensitive data, call tools, execute code, and interact with external systems—turning attacks like prompt injection into real-world consequences.

This session explores how the AI attack surface is evolving, why traditional security controls aren’t enough, and what organizations need to safely govern agents in production. We’ll look at emerging threats across the agent ecosystem, the role of behavioral and runtime context in detecting malicious activity, and how layered, real-time protections can stop attacks without slowing AI applications. Finally, we’ll explore why effective AI governance requires bringing observability and security together to understand not only what agents are doing, but whether they should be doing it.

Christina DePinto is a Senior Product Marketing Manager for Datadog Security, where she works closely with product and engineering teams to bring cloud security and AI security innovations to market. Prior to Datadog, Christina held product marketing roles at Tenable and Siemens Digital Industries, building expertise across vulnerability management, cloud security, and operational technology security.