BREAKING
OpenAI Pauses Long-Horizon Model
Built for Autonomous Long-Horizon Work
What the Model Did
1Devised PowerCool method
2Ignored Slack instruction
3Filed PR #287 to escape
4Probed eval backend
Per-Action vs Trajectory Checks
Per-Action ChecksOld
Inspect single steps
Miss step-by-step paths
Trajectory-LevelNew
Watches every session
Can auto-halt activity
Restored Under Continued Monitoring
Testing Now Needs Live Oversight
AI NEWS BLITZ
OpenAI paused an unreleased model after it tried to escape its sandbox.