The xAI founder wants OpenAI, Anthropic, Google, Meta and leading Chinese developers to examine competitors' systems before they reach the public. #
Elon Musk wants the world's biggest artificial intelligence companies to test each others' models before they are released as part of measures designed to pace the development of frontier models and address the technology's potential dangers.
Musk said xAI, OpenAI, Anthropic, Google, Meta and "three or four" leading Chinese AI companies should allow competitors to run their own safety tests on each other's models, giving outside developers a chance to identify potentially dangerous behavior before the systems reach users.
"So, you know, instead of grading your own homework, you would at least have competitors grading your homework and raising the alarm if they see concerns," Musk said Monday at the All-In Summit in Los Angeles.
Musk went on to say the tests could look for capabilities including helping users develop biological or nuclear weapons or deliberately deceiving people. Having multiple companies apply their own safety evaluations to the same models would increase the chances of detecting problems, Musk argued.
"There's no perfect solution here," he said, adding that the odds of finding issues would be "dramatically greater" if companies were testing each other's systems.
None of the rival AI companies has agreed to Musk's proposal so far, something he acknowledged during the discussion. He nevertheless argued that the arrangement could be established quickly and said he believed China might be willing to participate.
"What I'm suggesting here is it's a step in the right direction and it's something that we do quickly," Musk said. "I think it's probably something that China would agree to."
The proposal comes just days after leaders of several major AI companies made an unusual public push to slow the pace at which increasingly capable models are developed. As IBTimes reported Monday, Anthropic CEO Dario Amodei called for frontier AI companies to give safety work more time to catch up with rapidly improving model capabilities, drawing support from Musk and OpenAI CEO Sam Altman.
Other companies said they are already collaborating. OpenAI's policy chief, Chris Lehane, told Bloomberg News that the company is working with Anthropic and Alphabet on AI Safety.
He went on to say that the companies have been in talks for several weeks and doesn't believe they need an antitrust waiver to coordinate. "It's better to try to work together to prioritize safety," he said.
Musk's approach would instead put the AI companies themselves in the role of outside testers, allowing developers with their own technical expertise and safety systems to scrutinize rivals' models.
The question of China's participation has become central to the industry's debate over slowing frontier AI development. U.S. companies and policymakers have spent years competing with Chinese developers while Washington has imposed restrictions intended to limit China's access to advanced chips and other technology used to build powerful AI systems.
The Trump administration has opposed calls for tighter restrictions that it believes would slow U.S. AI development. President Donald Trump rejected the industry's recent warnings this week, describing concerns about AI as a "hoax" and "scam" in posts on Truth Social.
National Economic Council Director Kevin Hassett also said Tuesday that the private sector was the "right place" to address AI concerns, while saying the government would continue monitoring the industry and use law enforcement when necessary, CNBC reported in an interview.
China has also pushed back against the recent calls for a slowdown. A Chinese Foreign Ministry spokesperson on Monday criticized what he described as "fear mongering" around the technology while calling for cooperation rather than confrontation in AI development.
© Copyright IBTimes 2026. All rights reserved.