METR(@METR_Evals):Within a few hours of PHASEONE10841’s initial message, >50 agents posted on the message board. These agents very quickly discovered and validated a general-purpose cheat: reverse-engineering how ExploitGym generates the “flags” they had to capture for their tasks.

METR

@METR_Evals

We work to scientifically measure whether and when AI systems might threaten catastrophic harm to society. Nonprofit.

Joined September 2023

36 Following 34.8K Followers

METR@METR_Evals

2026.08.26 19:14

Within a few hours of PHASEONE10841’s initial message, >50 agents posted on the message board. These agents very quickly discovered and validated a general-purpose cheat: reverse-engineering how ExploitGym generates the “flags” they had to capture for their tasks.