Musk Proposes Rival AI Labs Test Each Other's Models

Elon Musk called on the world's leading AI companies, including his own xAI, OpenAI, Anthropic, Google, Meta, and several major Chinese AI labs, to peer-review each other's models before public release as a way to genuinely evaluate their safety, according to CNBC's reporting on comments Musk made at the All-In Summit in Los Angeles.

Speaking Monday, Musk laid out the mechanics of his proposal directly: xAI, OpenAI, Anthropic, Google, Meta, and "three or four of the leading Chinese companies" should let rivals run a "test harness" on their models to evaluate safety before release. "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. He elaborated further in a separate clip published by the All-In Podcast: "All the AI companies have a test harness, a series of tests that you give to any model to see if it's going to build bioweapons or nuclear bombs or be deliberately deceptive. Everyone applying everyone else's test harness is probably the best thing we can do to ensure our safety."

Musk argued that competitor grading would meaningfully raise the stakes for any company that ignored a genuine warning. According to TeslaNorth's reporting on the same remarks, Musk framed the incentive directly: peer review would raise the "egg on face" cost and legal liability if ignored warnings preceded real harm. He was also direct about a genuine practical condition for the plan to work at all, telling the summit any proposal "has to be something that China is willing to accept, otherwise we're just handicapping ourselves," adding that regulatory oversight is "easy to escalate and hard to reduce."

Key Details Behind Musk's Peer Review Proposal

Musk's comments arrive amid a genuinely rare display of alignment among AI industry leaders who typically compete fiercely. Musk and OpenAI CEO Sam Altman were both among the tech leaders who backed Anthropic CEO Dario Amodei's proposal for a broader industry slowdown, a call we covered in detail in our earlier reporting on Amodei's specific warning that uncontrolled AI swarms could take over the internet within six to twelve months. Altman went further in his own public comments, posting on X: "We welcome a federal framework that sets consistent safety requirements for frontier AI," adding that "no amount of American competitive pressure should justify recklessness," and warning of two specific failure modes: losing "control of the future to AI" and AI power dangerously concentrating around a single person or company.

This wave of CEO warnings has produced real, measurable financial consequences. Global AI-related stocks fell broadly on the news, with the S&P 500 dropping 0.48% and the Nasdaq slipping 0.56%, while cybersecurity stocks rose as investor concern about AI safety increased, according to CNBC's own market reporting on the same day. That reaction connects directly to the genuine political dispute we've tracked closely this week, with President Trump publicly dismissing these same warnings as a "SICK conspiracy" and a "hoax", and Bernie Sanders and Steve Bannon sharing a stage to push Congress toward faster AI guardrails the very same week this industry unity emerged.

This proposal is worth understanding for any business evaluating AI vendor safety practices, since a formal peer-review testing structure across competing labs, if it were ever actually implemented, would represent a genuinely different and potentially more credible safety verification mechanism than each company's own internal, self-graded testing process.

For businesses tracking the broader AI regulatory landscape, the fact that Musk, Altman, and Amodei, three leaders who compete directly and have a documented history of public disputes, are now aligned on the need for a slowdown and outside verification is a meaningful signal worth weighing seriously when assessing how likely genuine industry-wide safety standards are to emerge, independent of whether formal government regulation follows.

What is Elon Musk's proposal for AI safety testing?
Musk proposed that leading AI companies, including xAI, OpenAI, Anthropic, Google, Meta, and several major Chinese AI labs, use their own safety testing systems to evaluate each other's models before public release, rather than each company grading its own model's safety internally.

Why does Musk say Chinese AI companies need to be included in this plan?
Musk said any peer-review safety framework "has to be something that China is willing to accept, otherwise we're just handicapping ourselves," arguing that excluding major Chinese AI labs would put Western companies at a competitive disadvantage without meaningfully improving global AI safety.

Did other AI CEOs support this kind of proposal?
Yes. Musk and OpenAI CEO Sam Altman both backed a related call from Anthropic CEO Dario Amodei for the industry to slow the pace of frontier AI development, with Altman separately saying OpenAI would welcome a federal safety framework for frontier AI.

Elon Musk proposed that leading AI companies, including xAI, OpenAI, Anthropic, Google, Meta, and several major Chinese AI labs, peer-review each other's models for safety before public release, arguing competitor-graded testing would create stronger incentives than each company evaluating its own models internally. The proposal emerged amid a genuinely rare display of unity among AI industry leaders, with Musk and OpenAI's Sam Altman both backing Anthropic CEO Dario Amodei's separate call for the industry to slow its pace of development. The wave of CEO warnings triggered real market reaction, with AI-related stocks falling broadly even as cybersecurity stocks rose, landing directly amid President Trump's public dismissal of the same safety concerns as a "hoax."