Anthropic diverges from OpenAI and Google, opposing loosened state-level AI safety regulations

Odaily News: Anthropic is at odds with tech giants such as Google and OpenAI over a Massachusetts AI safety bill. The bill proposes requiring major AI developers to hire independent evaluators every four months to assess the potentially catastrophic risks posed by their AI models. If passed, the proposal could become the new standard for state-level AI regulation in the United States.

As the U.S. Congress continues to delay progress on federal AI regulation legislation, states are accelerating the rollout of their own rules. Anthropic's divergence from other tech giants on AI safety regulation also reflects the significant divisions within the industry over AI model risk assessment and the intensity of regulatory oversight. (The Information)