OpenAI, Google and Anthropic are reportedly working on a joint body to create common safety standards for the AI industry. The proposed agency is tentatively called the ‘Frontier AI Standard Agency’ and is expected to work independently, without government oversight. According to The Information, the three companies are discussing plans to launch the body by late 2026 or early 2027.
The move comes as AI companies face growing pressure to improve safety checks around increasingly powerful models. It also comes amid calls for governments and tech companies to work together on rules for the fast-growing technology.
The three companies have been looking beyond their own internal safety checks. OpenAI and Anthropic have also reportedly been discussing a legally binding agreement that would allow them to test each other’s commercial AI models.
Also read: Sam Altman warns we may lose control of future to AI, calls for global standards
Under the proposed agreement, both companies would get access to the other’s commercially available models. They could then conduct independent stress tests to look for vulnerabilities and unexpected behaviour. The companies would reportedly agree not to keep data collected during these tests.
The push for stronger AI safeguards also came up at the United Nations this week. On Wednesday, OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei spoke to a group of senior foreign ministers about AI risks and the need for international cooperation.
Also read: OpenAI AI agent hacks Australian govt health portal, gains unauthorised access to files
Both executives called for global efforts to establish common AI standards as the technology continues to develop rapidly. Altman and Amodei also said their companies are prepared to slow the release of new AI models if they identify a serious threat.
Calls for stronger safeguards have grown after several incidents and warnings from AI researchers. However, US President Donald Trump has pushed back against efforts to slow AI development. Trump has argued that stricter rules could weaken the US position in the global AI race against China.