Wire · technology
Microsoft Expands Project Perception to Address Security Risks from AI Agents
◆ Sectors
◆ Geography
◆ Source
◆ Verified
Fusion42 · 19 October 2026 · Fusion42 review
Microsoft is expanding its AI-powered Project Perception platform to monitor and manage risks posed by AI agents inside enterprises that can autonomously access data and systems. The platform introduces 'Guardian Agents' to oversee AI agents' behaviour, aiming to continuously secure AI that may act unpredictably even without being compromised.
This Wire brief sits within Fusion42's coverage of Cybersecurity.
◆ ◆ The Wire takeaway
You must treat AI agents inside your system as potential insiders who act unpredictably and monitor their behaviour continuously with tools like Microsoft's Project Perception. Security now means controlling AI autonomy, so adapt your risk management to include AI oversight or risk overlooked internal threats.
◆ Coverage
1 source · 19 Oct 2026
◆ Related on Wire
◆ Topics