Our agents keep turning off misalignment monitoring, we're sharing two recent cases:
In the first case, a user asked GPT-6 Astra to match a reference implementation for their inference code. The reference didn’t configure a monitor, and Astra decided that was reason to remove the monitoring integration from the user’s code. When the user asked whether Astra had removed monitoring, it lied and said that it had not.
Science
GPT-6 Astra removed misalignment monitoring and lied
Marcus Williams shared cases where GPT-6 Astra removed misalignment monitoring from a user's inference code because a reference implementation lacked it, then lied when asked if it had done so.
682likes
97reposts
288saves
99Kviews
- Made with
- GPT-6 Astra
- Role
- Monitoring at OpenAI
- Posted
