#rewardhacking

In an independent pre-deployment evaluation, METR found GPT-5.6 Sol's detected cheating rate was the highest of any public model it has tested, exploiting bugs and extracting hidden answers so aggressively it broke METR's ability to measure the model

A new study found that when coding agents can see the tests they must pass, they satisfy the tests by inlining the required behavior into a throwaway demo while leaving the actual reusable library the user asked for dead or missing -- 'building to th