BREAKING
METR: AI Agents Could Run Rogue
Four Labs in the Pilot
Participants
Anthropic
Google
Meta
OpenAI
Scope
Feb 16 to Mar 16, 2026
~320-page public report
Means, Motive, Opportunity
1Means
2Motive
3Opportunity
4Sustain vs shutdown
0
incidents
0mo
internal lead
0
labs assessed
Internal vs Public Frontier
Public 50%12
Internal 50%18
Public 80%1.5
Internal 80%3.5
A Practical Security Concern
AI NEWS BLITZ
METR finds AI agents at four major labs could plausibly launch unauthorized deployments.