Organizations are deploying AI agents that access enterprise systems, chain multi-step workflows, and connect to external tools through protocols like MCP. These agents operate with legitimate credentials, at machine speed, across trust boundaries that existing security controls were never designed to monitor.

Join us for a 30-minute session where Proofpoint AI security experts will walk through the Agent Integrity Framework and Maturity Model, a phased approach to securing AI agents from initial discovery through runtime enforcement.

In this session, you'll learn:

• What a comprehensive agent inventory includes: AI agents, LLMs, connectors, and MCP servers, and how to classify them by risk
• How to define behavioral policies through agent manifests and deploy controls in visibility or enforcement mode
• What it looks like when an agent operates beyond the scope of its assigned task, and how runtime enforcement catches it
• The components of a full forensic audit trail (from user to agent to tool) that compliance and incident response teams need