The New York Times: The A.I. Slowdown Debate Goes Global
By Andrew Ross Sorkin, Bernhard Warner, Sarah Kessler, Michael J. de la Merced, Niko Gallogly, Brian O’Keefe, Sri Muppidi and Ian Mount
Sept. 14, 2026, 8:01 a.m. ET
To pace or not to pace
It’s rare to see Dario Amodei of Anthropic and Sam Altman of OpenAI agree on much. So the world was struck when the leaders of those two artificial intelligence giants called for a slowdown — “pacing,” in Amodei’s words — in developing bleeding-edge models.
But many leaders in Silicon Valley, Washington and Beijing have pushed back.
“I won’t lie to you — there are real dangers,” Amodei told CBS News over the weekend about his call for a slowdown in developing the most advanced A.I. models.
He proposes giving independent A.I. evaluators “employee-like” access to companies’ models (somewhat like nuclear arms inspectors) and fostering international cooperation on safety standards.
Many of Amodei’s top rivals backed his calls for a global slowdown,including:
Altman, who wrote on social media that progress “should be slower than it otherwise could be.” Though OpenAI had reportedly worried about the potential antitrust concerns of a coordinated industry approach, Altman suggested that he wasn’t concerned about that.
Elon Musk of SpaceX, who wrote on X, “Dario is right.”
Demis Hassabis of Google, who wrote, “The direction is correctfor meeting this critical moment.”
The possible economic and financial consequences are huge.Goldman Sachs researchers estimated last month that A.I. accounts for about 1.8 percent of U.S. G.D.P., potentially rising to nearly 3 percent by 2028. A.I. fervor has also driven the stock markets’ astronomical gains in recent years.
A.I.-related stocks — including those of the chipmakers SK Hynix and Samsung Electronics, the Chinese lab Z.ai and the tech investor SoftBank — fell sharply in Asia overnight. S&P 500 and Nasdaq futures are down on Monday.
It’s notable that Amodei published his essay as Anthropic prepares for a blockbuster I.P.O., though critics have asked whether this actually serves as a cynical way to market the company as the safe A.I. player.
Amodei’s call has been criticized on multiple fronts:
President Trump, who has labeled American dominance of A.I. critical, said on Sunday that “I think you have a lot of negative forces” that keep bringing up A.I. guardrails.
David Sacks, the investor and former Trump administration A.I. czar, argued that calls for greater government regulation would hurt smaller rivals.
Clément Delangue of Hugging Face, whose company was hacked by OpenAI agents this year, wrote that safety issues “won’t be solved behind the closed doors of a handful of frontier labs.”
David Krueger, a prominent A.I. scientist and professor, called it “too little, too late.”
The big question is China. Amodei acknowledged that his plan wouldn’t really work unless the world’s other A.I. superpower agreed to a slowdown. So far, that appears unlikely: An editorial in the state-backed Global Times derided his essay as a “Cold War”-style attack on Beijing’s A.I. industry.
That said, some industry executives have told DealBook that, given accusations of how much Chinese models piggyback off American counterparts, any American slowdown would actually hamstring China’s A.I. companies too.
