Google, OpenAI, and Anthropic are reportedly working together to launch an independent organization dedicated to artificial intelligence safety. The proposed entity, tentatively titled the Standards Authority for Frontier AI (SAFA), is expected to be established by late 2026 or early 2027. According to a report from The Information, the companies are seeking to operate the body independently of government oversight.
The primary objective of SAFA will be to create practical benchmarks for the safety commitments made by the participating firms. Additionally, the organization aims to assist developers in evaluating advanced AI models before they are released for commercial use. The companies are currently considering whether SAFA should perform these model assessments internally.
Industry insiders have expressed concerns that existing government bodies, such as the United States Center for AI Standards and Innovation, may lack the necessary financial and technical resources to effectively inspect complex frontier AI systems. To lead the new initiative, the companies have approached former White House AI policy advisor Sriram Krishnan and former White House Office of Science and Technology Policy director Arati Prabhakar for the role of chief executive.
This collaborative effort follows a period of intense industry discussion regarding safety frameworks. An OpenAI spokesperson confirmed to CNBC approximately 10 days ago that the company is engaged in active talks with Google and Anthropic on the matter. The initiative gained significant traction following a proposal in July by Sir Demis Hassabis, the co-founder of Google DeepMind and chief scientist at Alphabet. Hassabis suggested the creation of a standards body similar to the Financial Industry Regulatory Authority (FINRA), functioning as a public-private partnership or a self-regulatory watchdog.
