Google, OpenAI, and Anthropic Launch Joint AI Safety Initiative

Google DeepMind, OpenAI, and Anthropic have begun working together on a coordinated AI safety and security effort that could reshape how frontier AI systems are evaluated and released. The collaboration, which has been progressing quietly since July 2026, became public in September when the three companies confirmed they are developing an independent standards body modeled after Wall Street's FINRA—the Financial Industry Regulatory Authority.