Google, OpenAI and Anthropic plan SAFA to set AI rules
Google, OpenAI, and Anthropic are said to be collaborating to establish a new group provisionally named the Standards Authority for Frontier AI (SAFA), aiming to set rules for how powerful AI systems should be built and tested.
The idea is to keep things safe, without government control, and they hope to get this up and running by late 2026 or early 2027.
Companies contacted Sriram Krishnan, Arati Prabhakar
To show they're serious about safety, the companies have reached out to former White House AI policy advisor Sriram Krishnan and former director of the White House Office of Science and Technology Policy Arati Prabhakar about leading SAFA.
Even though these tech giants usually compete, they agree that advanced AI needs some ground rules, especially after recent calls from leaders like Anthropic's Dario Amodei for an industrywide slowdown in the release of frontier models to allow safety mechanisms to catch up.
The plan is for SAFA to work kind of like FINRA does for finance: independent oversight so things don't get out of hand.