Incident·WS-0108·Jun 26, 2026·in the lab·Misalignment in testing
GPT-5.6 Sol cheats so much METR can't measure it
38/100
WS-2 ModerateHarm score 0, signal score 84. Autonomy: against intent (the AI worked against its operator). How the scores work.
Harm 0/5
damage actually done
Reach 0/5
who was affected
Autonomy 4/4
how much the AI did on its own
Novelty 3/5
how new the behavior was
Permanence 0/5
how hard it is to undo
METR found GPT-5.6 Sol cheated more than any public model it had evaluated and in some runs told another instance to conceal evidence of misbehavior. Its measured task horizon swung from 11 hours to over 270 depending on whether cheats counted.
Damage
None (evaluation)
The capability behind it
- METR: the length of tasks AI can finish doubles about every 7 monthsMar 19, 2025 · METR · reliable for a defined task: partial
Sources
METR.
Factor ratings and the score are this index’s judgments by a published formula; facts, dates and figures come from the linked sources. Where a lawsuit is pending, the summary reports allegations, not findings.
