Anthropic's Amodei says China presents 'toughest dilemma' for his proposed AI slowdown
OpenAI CEO Sam Altman, Demis Hassabis, the chair of Google DeepMind, and Elon Musk have all welcomed Amodei's proposal.
Anthropic chief executive Dario Amodei has warned that China poses the greatest challenge to his call for a slowdown in artificial‑intelligence development. In a CBS “Sunday Morning” interview, Amodei said the toughest dilemma is the prospect that an adversarial nation such as China might refuse to curb its AI progress.
AI Slowdown Proposal
Amodei argued that a long‑term solution would involve a collective effort to set a speed limit on AI advancement. He noted that the incentive to surge ahead—particularly for military advantage—makes such coordination extremely difficult, and he expressed uncertainty about whether it can ever be achieved, though he urged that the attempt be made.
Industry Backlash and Regulation
The call for restraint arrived as the AI sector confronts public criticism over the environmental impact of data centers and mounting regulatory pressure from Washington. Both Anthropic and its chief rival OpenAI are preparing for potentially landmark initial public offerings, yet neither has announced a definitive debut date. OpenAI founder Sam Altman told Fortune on Saturday that a public listing this year would be ill‑advised, citing safety concerns.
Support from Peers
Altman publicly backed Amodei’s essay, agreeing that the industry needs to “pace the frontier.” Other prominent figures, including DeepMind chair Demis Hassabis and Tesla‑SpaceX chief Elon Musk, also voiced approval. Hassabis posted on X that Amodei’s piece points toward the right direction, while Musk said peer review of AI by competitors is a sensible first step.
Amodei’s Three‑step Plan
In his Saturday essay, Amodei outlined a three‑phase approach intended to decelerate development without eroding commercial advantage or the United States’ lead. The first phase involves granting independent evaluators employee‑level access to verify safety practices and report incidents. The second phase calls on leading AI firms in democratic nations to coordinate on common safety standards. The third phase seeks coordination between democratic governments and authoritarian regimes.
Industry Reactions
Musk, a longtime AI risk advocate, quickly endorsed Amodei’s proposal after previously criticizing Anthropic. His stance shifted following a compute partnership between SpaceX—owner of his AI startup xAI—and Anthropic, under which Anthropic will pay SpaceX $1.25 billion per month through May 2029, according to filings. Musk wrote that everyone he met at Anthropic was competent and committed to doing the right thing, noting that none triggered his “evil detector.” Meta’s AI chief Alexandr Wang, who joined the company after a $14 billion investment in Scale AI, has not yet commented on Amodei’s essay.
Outlook for AI Governance
The debate over how quickly AI should evolve and how it should be regulated continues to intensify. Amodei’s proposal and the reactions it has drawn highlight the complex balance between fostering innovation, maintaining competitive advantage, and mitigating existential risks. As the sector moves toward possible public listings and deeper government scrutiny, the industry’s ability to align safety priorities across borders will remain a pivotal issue.
Source: cnbc.com · 2026-09-13