Anthropic has onboarded six METR evaluators with company hardware, office space and internal system access comparable to its own risk teams. The nonprofit can publish findings on the AI lab's risks and practices without Anthropic's approval, though limited redactions remain possible. METR sits in the top tier of a new three-tier evaluator system Anthropic announced Oct. 15.
Anthropic has onboarded six METR evaluators with company hardware, office space and internal system access comparable to its own risk teams. The nonprofit can publish findings on the AI lab's risks and practices without Anthropic's approval, though limited redactions remain possible. METR sits in the top tier of a new three-tier evaluator system Anthropic announced Oct. 15.This article was originally published by Webpronews.
Read Full Story on Source Site →
