Over 1,200 AI employees urge US government to pump the brakes on frontier development

Over 1,224 employees from OpenAI, Anthropic, Google (including DeepMind), and Meta have signed an open letter titled “Pacing the Frontier,” urging the US government to support international governance efforts that would deliberately moderate the pace of frontier AI development. What the letter actually says The petition argues that AI capabilities are advancing faster than society’s ability to understand or control them, and that some kind of coordinated mechanism is needed to buy humanity time to catch up. Anthropic CEO Dario Amodei and OpenAI Chief Scientist Jakub Pachocki are among the notable names attached to the letter. The letter specifically highlights the problem of competitive pressure. No single company or country can unilaterally slow down without risking falling behind rivals. That’s the whole reason the signatories are calling for verifiable coordination, essentially an international framework where everyone agrees to ease off the accelerator together. One of the core concerns raised in the letter involves recursive self-improvement trends in AI — the worry that AI systems are getting close to being able to improve themselves, which could trigger a feedback loop where...