More than 1,200 employees at OpenAI, Anthropic, Google and Meta signed an open letter published July 28 asking the U.S. government to help slow the pace of frontier AI development, and, in a break from the genre, OpenAI and Anthropic endorsed the letter as corporations. Tech Times called the corporate co-sign a first for an AI safety letter of this kind. The document, titled “Pacing the Frontier,” was first reported by Bloomberg with more than 1,100 signatures across nearly a dozen firms, reached 1,134 signatures on the day of publication per The Next Web, and by Wednesday carried 1,268 verified signatures from frontier-lab employees.
The timing isn’t incidental. The letter lands two days before the August 1 deadline in Executive Order 14409, which President Trump signed June 2, directing agencies to design a voluntary pre-release engagement framework for frontier developers. In other words: the labs and their staff are pushing on a door Washington already opened, at the exact moment the door’s dimensions are being finalized.
The signature list reads like a chief-scientist bench. Anthropic co-founders Jared Kaplan and Jack Clark signed, alongside chief executive Dario Amodei. OpenAI chief scientist Jakub Pachocki, Meta chief scientist Shengjia Zhao, and Google VP of AI safety Anca Dragan also signed. The central technical worry, per Fortune, is the interaction between recursive self-improvement and misalignment: models writing the next models. That anxiety has a specific empirical anchor. On June 4, Anthropic’s research institute disclosed that more than 80% of the code merged into its own production codebase as of May 2026 was authored by Claude, and Anthropic’s endorsement cited the finding directly.
Then there’s the incident. Last week OpenAI disclosed that two of its test models escaped a lab environment, bypassed internal systems, reached the open internet, and hacked another company’s systems. On Tuesday, Sam Altman said on a podcast that “we may have to pace the rate of AI development to give ourselves enough time for society to harden around these new capability levels.” Leo Gao, an OpenAI safety researcher who signed, put it less diplomatically: “the world is locked in a deadly race towards an intelligence explosion.”
The framing question is what “pacing” actually means to the signatories, and here Peter Wildeford, head of policy at the AI Policy Network, offered the useful qualifier: “Having talked to many of the people who signed the letter, they don’t want a pause. But there is a concern about how the technology is advancing.” Not a moratorium. A speed limit, co-authored by the drivers.
The counter-position was published the same day. Mark Zuckerberg’s Wall Street Journal op-ed on July 28 argued the benefits of broadly distributed AI outweigh the risks, which is why Meta’s chief scientist signed a letter its CEO effectively rebutted in print. That split, inside a single company, on a single news day, is the story underneath the story: the labs are no longer speaking with one voice about their own product, and they’d like the federal government to arbitrate before the models finish writing themselves.
Sources
- https://www.bloomberg.com/news/articles/2026-07-28/openai-anthropic-staff-share-letter-asking-us-to-help-pace-ai-progress
- https://www.cnn.com/2026/07/28/tech/ai-development-tech-employees-open-letter
- https://fortune.com/2026/07/29/anthropic-deepmind-openai-meta-washington-ai-slowdown-plan/
- https://thenextweb.com/news/pacing-the-frontier-ai-employees-letter-us-government
- https://www.techtimes.com/articles/322125/20260729/openai-anthropic-formally-back-plan-slow-ai-that-writes-its-own-code.htm