Hassabis' AI Standards Idea Gets Support, What's Next?

Google DeepMind CEO Demis Hassabis proposed creating a U.S.-led AI standards body modeled after FINRA. The organization would create quarterly-updated benchmarks for frontier models, evaluating risks in cybersecurity, biology, and deceptive AI. Hassabis hopes to launch the body by year's end after talks with the White House, European officials, and AI labs.