Anthropic models are far and away safer than any other model. They are the only ... | Hacker News

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

		conception 3 months ago \| parent \| context \| favorite \| on: Claude's new constitution Anthropic models are far and away safer than any other model. They are the only ones really taking AI safety seriously. Dismissing it as PR ignores their entire corpus of work in this area.

viccis 3 months ago [–]

By what measure? What's "safe"?

conception 3 months ago | [–]

https://crfm.stanford.edu/helm/air-bench/latest/#/leaderboar...

This isn’t the gotcha question you think it is. AI safety is being defined and measured.

viccis 3 months ago | | [–]

Cool, another metric to game like they do the other ones.

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