Anthropic is letting METR researchers look under the hood with employee-level access

via TechCrunch · 3 views · 1d ago

Anthropic just gave researchers unprecedented access. It's a huge move for transparency in AI development. The frontier model maker is letting METR researchers look directly "under the hood" of their systems, granting them employee-level access. What’s more, METR will publish its own independent assessments of Anthropic’s risks, incidents, and practices. Crucially, Anthropic has explicitly given up the right to block these findings. This isn't just a peek behind the curtain; it’s a commitment to external scrutiny, setting a new bar for how AI developers might be held accountable. Expect more detailed insights into advanced AI safety.

← Back to the feed