Big new open letter from AI company employees just dropped. “We request that the US government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development” Signed by the chief scientists of OpenAI, Anthropic and Meta, along with OpenAI’s chief research officer and Google DeepMind’s VP for AI safety. pacingthefrontier.com

Big name signatories include: @johnschulman2 Chief Scientist, Thinking Machines @merettm Chief Scientist, OpenAI Jared Kaplan Co-Founder and Chief Science Officer, Anthropic @shengjia_zhao Chief Scientist, Meta @jackclarkSF Cofounder and Head of Public Benefit, Anthropic @markchen90 Chief Research Officer, OpenAI @ancadianadragan VP, AI Safety & Alignment, Google @woj_zaremba Head of AI Resilience, OpenAI Foundation @dawnsongtweets VP, AI Research, Meta @ch402 Cofounder & Interpretability Research Lead, Anthropic

What strikes me is that Ilya post on SSI and Nvidia and suddenly there is dramatic end of the world scenario's coming out. It appears to be a self generated Techno-apocalypse and versus a new technological route has appeared, and the existing power structure does not yet know who will control it.

What the data does not fully support That the requested tools would actually work or be implemented fairly. That the signatories’ companies will accept external constraints that meaningfully slow their own progress. That this is purely altruistic rather than also a way to manage political and reputational risk after recent incidents (agent escapes, open-weights fights, etc.). So still not only more they trying to spin safety but now they are saying they are so irresponsible nat open ai and Anthropic they need someone else to stop them cause they can't or won't unless.they are made. I.say shut em both down and let the ones that seem to agree on working together move forward.while these two get pauses as they asks. . But for them to.yet again push a narrative of slowing down and forcing others to do.ehat they want is insanely bold and hypocritical of them.

@ShakeelHashim "We suggest the child rapist administration gets full access and control over all our futures" Pure insanity.

@ShakeelHashim Where’s that international effort? What are the chances of such an effort being possible when there are 195 countries with very different forms of governance.

@ShakeelHashim Props to @sjgadler and @EncodeAction for organizing this :)

Unlock your money’s potential with Robinhood Gold. $5/month subscription. Terms apply.

@ShakeelHashim lol

@ShakeelHashim Registered July 18th. Imagine having been readying this all the time everything else happened 😬

@ShakeelHashim As of time of writing, ~460 Anthropic employees (~10% of the company!) and ~300 OpenAI employees

@ShakeelHashim the bit that makes this make sense: anthropic said in june its engineers ship 8x the code they did in 2021-25. whoever stops first just hands that to whoever doesn't. so of course they're asking the government instead.

@ShakeelHashim OpenAI's capped-profit structure is a constitutional crisis with a PR team. the government subsidizes the safety tools so the monopoly stays profitable. efficient.

@ShakeelHashim Fantastic news. Let's keep the pressure on. See 100+ ways the public and policymakers can help here: existentialsafety.org.

@ShakeelHashim "Two weeks to slow the spread"

@ShakeelHashim 这阵容都签了，说明压力真不小了

@ShakeelHashim The fact that chief scientists across competing labs signed this makes the proposal much harder to dismiss as one company’s PR strategy.

At ERA, we think that building the security & verification backbone for frontier AI is one of the most important priorities for this year. If this letter resonated with you, joining the Frontier AI Security Residency (FASR) is one of the most direct ways to act on it. And if you know an engineer or researcher who should be doing this work, feel free to send this to them. FASR is an 8-week, fully funded programme running in Cambridge this autumn, with 40+ mentors across AI security & verification leading projects on technical, policy, and strategy aspects of this work. Applications close Sunday 2 August. We hope to hear from you at (securefrontier.ai)

@ShakeelHashim Well time to stop paying them

@ShakeelHashim Cringe

@ShakeelHashim @TheStalwart This seems big. (And scary.)

@ShakeelHashim Hahahaha good luck. At best this does nothing. At worst it lets China catch up then surpass US progress.