News · curated 3 Oct 2026
Set up AI agent runtime protection with Microsoft Defender for Endpoint (Preview) - Microsoft Defender for Endpoint | Microsoft Learn
First reported microsoft.com
Coverage timeline
Single-source advisory — first reported, latest, and curated coincide.
Why it matters
Microsoft Defender for Endpoint's runtime protection targets prompt injection against local AI agents that can read files, invoke tools, and run commands, giving defenders an endpoint-level control against agent hijacking.
Microsoft documentation describes AI agent runtime protection in Microsoft Defender for Endpoint (Preview), a defensive capability that detects prompt injection at the device level against local AI agents running with user privileges. The feature can block or audit an agent's action before it acts on malicious instructions hidden in content the agent reads, and the article covers enabling, deploying, and investigating detections.