Amodei Urges AI Slowdown, Cites China as 'Toughest Dilemma' — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Dario Amodei published an essay Saturday urging AI companies to slow how quickly they improve their most advanced models, telling CBS "Sunday Morning" the "toughest dilemma" is what happens if adversarial nations like China refuse to follow suit.
- Amodei's three-step plan calls for third-party evaluators with employee-level access to verify safety — a step Anthropic has "unilaterally" adopted — followed by coordination among democratic-country AI firms and ultimately with authoritarian governments.
- Anthropic researcher Jacob Coxon announced he quit his job, publicly warning that Anthropic and rival OpenAI are "gambling with our lives" with their frontier-model race.
- Sam Altman said OpenAI will likely not go public this year, citing safety concerns, while endorsing Amodei's call to "pace the frontier" — a notable public alignment between the two rival CEOs.
- Elon Musk and Demis Hassabis both backed the proposal on X, with Musk's endorsement arriving after SpaceX struck a $1.25 billion-per-month compute deal with Anthropic running through May 2029.
- David Sacks, Trump's former AI and crypto czar, offered a more conditional endorsement, telling OpenAI and Anthropic to "stop pretending you need anyone else's permission" and warning that seeking a regulatory framework would look like "blackmail of the public."
Why it matters: The pushback from Sacks — a Trump-aligned voice — reframes Amodei's slowdown campaign as commercial self-interest rather than altruism, given that both Anthropic and OpenAI face 'massive product-liability exposure' and are gearing up for potentially historic IPOs. The China angle is the real binding constraint: Amodei himself admits he doesn't know if a multilateral speed limit is possible given the military advantage at stake.
Ask SkimNews
