US Rejects French Plan for G7 AI Safety Group

Politico reports that the US government rejected a French proposal for the G7 to coordinate advanced AI safety work. A White House official described the proposed group as poorly defined and said it could create unnecessary bureaucracy.

French digital minister Anne Le Hénanff said meetings in September and October sought commitments to a high level group that would test frontier AI models and share information about risks and incidents. She said the US wanted to complete its own assessment before supporting a joint statement.

France issued a chair's summary after an October 20 virtual meeting of G7 technology ministers, rather than a statement endorsed by all members. Le Hénanff proposed resuming talks in December and said France would continue the work through its national AI evaluation and research institutions.