Google DeepMind CEO Calls for Global AI Oversight Agency

Demis Hassabis, CEO of Google DeepMind, is urging the U.S. government to create a global AI oversight agency. He says it should act like a 'brake' for risky AI models, able to slow down releases when needed.

Hassabis suggests the agency should be modeled after the U.S. Financial Industry Regulatory Authority (FINRA), bringing in independent experts and open-source reps to check AI models before launch. He's been talking with U.S. officials, other AI labs, and European leaders for months, hoping the agency can start working by the end of this year.