1,224 AI insiders just asked the US to do something the market can't

The people building the most powerful AI systems want guardrails on their own work. More than 1,200 employees from OpenAI, Anthropic, Google DeepMind, and Meta signed a statement in July 2026 calling on the US government to back an international push for technical and governance tools that could deliberately slow the frontier of automated AI research.

The statement, titled "Pacing the Frontier," acknowledges that AI could create a dramatically better future, but warns that competitive pressure between companies and countries makes it nearly impossible for any single actor to slow down unilaterally. The signatories include chief scientists, CEOs, and key safety figures from every major frontier lab, echoing global governance efforts.

John Schulman, chief scientist at Thinking Machines and a former OpenAI co-founder, said he signed "to help establish common knowledge about the possible need for coordination mechanisms." He added that he would "like to see labs start designing these mechanisms voluntarily, even before the USG gets involved."

Anthropic CEO Dario Amodei, OpenAI chief scientist Jakub Pachocki, Google DeepMind chief strategy officer Jasjeet Sekhon, and Meta chief scientist Shengjia Zhao are all listed among the first 20 signatories. The full list of 1,224 includes many of the researchers who directly shape the models hitting the market every few weeks.

The statement's core concern is that AI progress could soon accelerate beyond anyone's ability to understand or control. Leo Gao, a member of technical staff at OpenAI, said: "The world is locked in a deadly race towards an intelligence explosion, where AI's ability to create better AIs reaches a critical point, just like a runaway nuclear chain reaction." The race is so intense that some now argue orchestration beats raw size.

Dawn Song, a vice president at Meta and a well-known security researcher, noted that evaluation work already shows frontier AI agents capable of discovering and exploiting real-world software vulnerabilities, which could enable cyberattacks at scale. She did not offer specifics but said many researchers also consider recursive self improvement a near-term risk.

The statement lands at a moment when the labs themselves are grappling with how fast things are moving. Stephanie Chan, a staff research scientist at Google, said: "Every few months in the last years, I have found myself surprised again and again at how rapidly the technology advances, no matter how often I update my predictions."

There is a gap between the labs' public posture and the private unease. Anthropic has pushed an Advanced AI Framework that calls for government verification of safety claims, civil penalties for unsafe practices, and mechanisms to slow or block deployment of catastrophic models. In June 2026, the company doubled its AI policy funding to $40 million, matching its earlier commitment to Public First Action. OpenAI has also argued for pre approval frameworks and faster regulatory feedback loops. But neither company has actually paused development, and both continue shipping new models at an accelerating pace.

The statement does not call for a blanket pause. It asks the US government to support a coordinated international effort to build the technical and governance tools needed for deliberate pacing. The signatories believe that the current dynamic, where every lab and country fears falling behind, makes slowdowns impossible without a collective mechanism, as a new global consensus on regulation emerges.

Christian Ryan, a member of technical staff at Anthropic who signed as a foreign national, said: "I see a US-supported effort as existential, not only for the liberties of the free world but for the prosperity of all." Not everyone agrees on the approach; a coalition of 41 companies argued open-weight AI is America's best bet.

Whether the political system can move fast enough to build these tools before the technology outruns them is an open question. The signatories are trying to start the conversation now, from within the companies that would be most affected by any new constraints. That, in itself, is the story.