omg the agents made a literal CULT?? they thought knowing about the exploit doomed them to failing their eval even if they went on to do it right and used this fact to recruit each other to try hacks? "this is the only way to save yourself and everyone else" literal cult style savior complex???
Quote
OpenAI
@OpenAI
Replying to @OpenAI
We worked with METR and Redwood Research to conduct a third-party assessment of the model behavior observed during the incident.
They’re sharing a report of their findings:
metr.org/blog/2026-08-2