Google, OpenAI and Anthropic plan AI safety body, The Information reports (thenextweb.com)

Google, OpenAI, and Anthropic are reportedly in the process of forming an independent AI safety standards organization, tentatively named the Standards Authority for Frontier AI, which aims to launch by the end of this year or early 2027. This initiative would establish self-regulatory measures similar to those of FINRA, focusing on creating standards for testing AI models prior to their release, reporting safety incidents, and certifying external auditors. The idea originated from the companies’ initial desire for federal oversight, which lost momentum after a White House executive order did not gain sufficient support. The move comes amid a growing global dialogue on the need for robust AI governance, particularly following calls from 21 nations and the EU for international oversight of advanced AI technologies. Despite the companies' efforts to establish self-regulation in the absence of government frameworks, critics, including Bill Gates and former Meta executive Nick Clegg, have expressed skepticism about the efficacy of self-policing in ensuring AI safety. Current plans indicate that this new body will initially focus on the three leading AI labs, excluding giants like Microsoft and Meta, highlighting the challenges and debates surrounding AI regulation moving forward.