On September 29, 2026, tech executives from OpenAI, Anthropic, Meta, Google, and Nvidia signed a voluntary, morally binding agreement at the White House establishing baseline AI safety standards.