Google, OpenAI, and Anthropic plan AI safety body by early 2027

Google, OpenAI, and Anthropic are collaborating to establish an independent AI safety organization, provisionally named the Standards Authority for Frontier AI, with a target launch by late 2026 or early 2027. The planned body aims to define practical benchmarks for the companies' public safety pledges and support developers evaluating cutting-edge models before deployment. Former White House AI policy advisor Sriram Krishnan and former OSTP director Arati Prabhakar have been approached for leadership roles, with discussions ongoing about conducting in-house model assessments. The initiative follows growing industry pressure, including a July proposal from Google DeepMind's Demis Hassabis for a US-backed standards body and an earlier essay from Anthropic CEO Dario Amodei calling for a slowdown in frontier model releases to let safety mechanisms catch up.