OpenAI, Google and Anthropic are reportedly working on a new industry body focused on AI safety standards. The proposed organization could launch by the end of 2026 or early 2027, according to The Information. The three companies are discussing an industry-led standards organization that would operate without direct government oversight.
The proposed body is tentatively being considered under the name Standards Authority for Frontier AI. The discussions come as leading AI companies face growing pressure to improve testing and auditing for increasingly capable models. OpenAI already confirmed that it is working with Anthropic and Google on AI safety issues for several weeks.
The proposed body would focus on common standards for evaluating frontier AI systems. Representatives from the three companies have reportedly been meeting regularly to discuss how such a system could work. OpenAI and Anthropic are also reportedly discussing an agreement that would allow them to test each other’s commercial AI models. The planned tests would look for vulnerabilities and unexpected behaviour.
Under the reported arrangement, the companies would have access to each other’s commercially available models for independent stress testing. Data collected during those tests would reportedly not be retained. OpenAI global policy chief Chris Lehane said earlier this month that discussions with Anthropic and Google had already been underway for several weeks.
The proposed standards body could be announced by the end of 2026 or during early 2027. The final structure, leadership and scope have not yet been settled, according to reports. The talks also follow recent public calls for stronger cooperation on AI safety. Anthropic CEO Dario Amodei called for greater coordination among AI companies and independent evaluation of advanced models.
OpenAI CEO Sam Altman also backed industry-led testing and auditing efforts. The companies are therefore exploring a voluntary industry framework while governments continue developing their own approaches to AI regulation. The proposed organization would focus on technical standards rather than replacing government rules.
Also Read: Google Launches Gemini 3.8 Live Avatar for Real-Time AI Conversations