The heads of the biggest AI labs — Google, Anthropic, Meta, OpenAI, SpaceXAI, and Nvidia — went to Washington and signed the 'Joint Commitment on Frontier Responsibilities,' promising to develop their own models safely.