AI Development
Microsoft expands Project Perception for agentic insider controls
Editorial Analysis
Microsoft released new security capabilities in Microsoft 365 E7 and Project Perception that treat AI agents as potential insider risks. The system discovers agents, assigns identities and policies, computes dynamic risk scores, and deploys Guardian Agents to supervise and constrain actions in real time. The update directly addresses enterprise concerns about unauthorized or high-risk agent behavior as autonomous systems proliferate inside organizations.
At a Glance
Date
October 19, 2026
Importance
Medium
3/5
Category
deployment
Axis of Change
Deployment Expansion
Organizations
Models Affected
Sources