AI Leaders' Slowdown Pact: Safety Move or Cartel? — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- AI leaders — OpenAI's Sam Altman, Anthropic's Dario Amodei, Google DeepMind's Demis Hassabis, and SpaceX's Elon Musk — loosely agreed over the weekend to slow development, signing on to Amodei's three-step proposal for third-party auditors, domestic lab regulation, and a global slowdown agreement.
- Anthropic researcher Jacob Coxon resigned and posted a public letter saying the people building AI "earnestly believe that it could kill us all by the end of the decade" and that neither OpenAI nor Anthropic is "acting responsibly" — the letter has been viewed more than 170 million times on X.
- Daniel Kokotajlo, ex-OpenAI and now head of the AI Futures Project, said CEOs are "bowing to that pressure and also claiming credit for it, [although] not rightfully," pointing to a July letter signed by more than 1,000 AI lab employees after the OpenAI-Hugging Face incident that already called for a slowdown.
- NYU's Nick Reese compared the move to social media platforms a decade ago calling for regulation to get ahead of unfavorable laws, saying "the hammer may not come in this administration" but could arrive under a future Democratic administration.
- President Trump posted Monday that "the only control or 'guardrails' that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT" and called Nvidia CEO Jensen Huang on speakerphone during a conference to call AI concerns a "hoax."
- Chinese Foreign Ministry spokesperson Guo Jiakun pushed back on the slowdown calls Monday, calling them "fearmongering," while experts like Redwood Research's Buck Shlegeris argued international coordination on AI would not be unprecedented, comparing it to US-Russia nuclear de-proliferation.
- Sacha Haworth of the Tech Oversight Project warned that any voluntary framework is regulatory capture and "we should not be letting the foxes run the henhouse," while Kokotajlo's AI Futures Project proposed AI labs give auditors access to their compute budgets and cut their own research budgets to let competitors catch up.
Why it matters: Because the Trump administration is actively hostile to AI regulation and China is refusing to slow down, this voluntary CEO pact is effectively the only governance mechanism in play — meaning the frontier labs are simultaneously the regulated. If the agreement holds, it locks in the competitive moat of incumbents against open-source developers; if it collapses, the trigger (Coxon's letter, 170M views) suggests public pressure will keep escalating until something gives.
Ask SkimNews
