No it happened during an ExploitBench eval. But I believe the same model already... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		cubefox 7 days ago \| parent \| context \| favorite \| on: GPT-6 Astra No it happened during an ExploitBench eval. But I believe the same model already cheated during training which wasn't detected until later.
		help

estearum 7 days ago [–]

Ah yes it was that a model in training found the Artifactory board, which was then more fully exploited during the ExploitGym eval

cubefox 7 days ago | [–]

Ah, ExploitGym. Not ExploitBench.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact