Dario Amodei published an essay on a Saturday morning, and by Monday the president of the United States was posting in all caps about it.
The Anthropic CEO’s essay, titled “We Must Pace the Frontier,” laid out why AI development should be slowed down. What followed was a few days of AI leaders and politicians lining up on either side of it, and the split doesn’t fall where you’d expect.
The essay itself is narrower than the reaction suggests
Amodei outlined three steps for pacing AI development. First, embedded third-party evaluators that can verify whether a company is adhering to safety practices and commitments and report incidents. Second, coordination between frontier AI companies in democratic countries on standards and limits. Third, global coordination between democratic governments and authoritarian governments on pacing AI development “to the extent this is possible.”
Anthropic is “unilaterally committing to the first of these steps,” Amodei said.
Read the fine print and the ask shrinks. Pacing “does not mean halting model training or technical progress, ensuring companies take adequate time to align and safeguard their models, and for third party evaluators to confirm this,” Amodei said.
OpenAI agreed faster than anyone expected
“I agree with Dario that we need to pace the frontier,” OpenAI CEO Sam Altman said shortly after Amodei posted about the essay on X. “This has been a primary topic of discussions we’ve had at OpenAI in recent weeks.” Altman committed to having independent evaluators at the company too, calling it a “great idea.”
He went further on Sunday. OpenAI welcomes “a federal framework that sets consistent safety requirements for frontier AI,” Altman said, adding that the company hopes others “learn from our approaches and propose their own; we think shared standards for misalignment, monitoring, and safety will lead to better outcomes.”
And he echoed Amodei’s hedge almost word for word. “When we talk about “pacing”, we do not mean “stopping”,” Altman said, noting that progress should be “slower than it otherwise could be.”
Altman also named two ways he thinks AI progress “could go very badly”: losing “control of the future to AI” and ending up “in a world with too much concentration of power,” which he said can be avoided by “walking a narrow middle path.”
Hassabis and Musk signed on with a sentence each
“Dario’s essay points towards the right path forward,” Google DeepMind co-founder and chair Demis Hassabis said in reply to the essay. “The details need working through, but the direction is correct for meeting this critical moment.” Hassabis pointed to a framework for an AI standards body he published in July.
SpaceX CEO Elon Musk was briefer. “Dario is right,” Musk said, reposting the essay.
Then Trump woke up
In a rambling Monday morning Truth Social post, the president pushed back on the idea of an AI slowdown:
“The only control or “guardrails” that AI needs is a STRONG AND SMART (High IQ!) PRESIDENT, and the U.S.A. has that, in spades! The Trump Administration has stopped AI “people” from doing bad, or potentially bad, “things,” like Dario (Anthropic!), who is now pretending to be a “perfect little angel” – and we will continue to do so! We already have tremendous CRIMINAL and REGULATORY power over these companies! There is a SICK conspiracy going on against AI and Data Centers, and the only one that is happy about it is China. WHOEVER WINS AI, WINS! We are leading China, and all others, and will continue to do so. Conspiracy Theorists, Treasonists, Traitors, and Leakers, BEWARE! Thank you for your attention to this matter! President DONALD J. TRUMP”
The suspicion from Washington is about motive
Vice President JD Vance didn’t argue the safety case. He questioned why the companies want rules at all.
It’s a “little bit weird” that frontier AI tech companies are “kind of coming to the government and begging the government to regulate them,” Vance said, adding that it “feels a little bit to me like a bit of a trojan horse.” He pointed to the president saying that “we know the Chinese are engaged in this long term AI race against the United States.”
That’s the tell. When the two biggest labs both ask for a federal framework in the same weekend, the people who’d have to write it start wondering who the rules would lock out.
The House Speaker says he’d meet tomorrow
In a Sunday interview with Jake Tapper on CNN, House Speaker Mike Johnson said that “I think the leaders and the providers of all these platforms… They need to come together in a meeting with us and sit down and figure out what the right balance is. We cannot put a moratorium on this because China will overlap us. And that’s the challenge. It’s national security balanced with the immediate security of making sure the models are safe.”
It’s unclear if or when this meeting might actually happen. Johnson said he would do it “tomorrow.”
The strangest alignment of the week
Senator Bernie Sanders said he agrees with Wall Street Journal columnist and former Ronald Reagan speechwriter Peggy Noonan, who published an article titled “Pause AI for Humanity’s Sake.”
A democratic socialist senator and a Reagan speechwriter landing in the same place, while the CEOs of the two leading labs ask to be regulated and the White House tells them no. Whatever coalition forms around this, it won’t run along party lines.
