Our Advanced AI Framework sets out how governments should prepare for and prevent catastrophic risks from frontier AI systems. The government should have the authority to block or revoke the release of unsafe models, and invest in societal resilience.

→ View original post on X — @anthropicai