Hassabis calls for US-led global AI safety authority with enforcement power

The detail worth sitting with is the US-as-steward framing. Proposing American leadership of a global body isn't a neutral governance preference, it's a competitive moat argument: a US-anchored watchdog would almost certainly reflect the compliance posture of US frontier labs, disadvantaging Chinese competitors and smaller international players who had no seat at the table when the standards were written.

Modelwire has no prior coverage to anchor this to directly, so it sits largely on its own for now. The broader context it belongs to is the ongoing debate among frontier labs about whether proactive regulatory engagement is a defensive strategy or a genuine safety commitment. Hassabis's call echoes arguments that have circulated since the early post-GPT-4 governance push, where large labs began treating regulatory frameworks as barriers to entry rather than constraints on themselves.

Watch whether any other frontier lab CEO (specifically Anthropic or OpenAI leadership) publicly endorses the US-steward framing within the next 60 days. Coordinated support would confirm this is an industry-aligned lobbying position rather than a DeepMind-specific stance.