AI executives and politicians debate pacing frontier AI development after Anthropic CEO’s proposal

Anthropic CEO Dario Amodei argued in a public essay that AI development should be paced to ensure adequate safety alignment and third-party verification. He proposed three steps: embedding third-party evaluators to verify adherence to safety practices, coordinating standards among frontier AI companies in democratic countries, and fostering global coordination between democratic and authoritarian governments where possible. Amodei stated that Anthropic is unilaterally committing to the first step—embedded third-party evaluators—and emphasized that pacing does not mean halting technical progress.

OpenAI CEO Sam Altman publicly agreed with Amodei’s call to pace frontier AI development and said OpenAI has been discussing the topic internally in recent weeks. Altman committed to having independent evaluators at OpenAI and endorsed a federal framework that sets consistent safety requirements for frontier AI. He cautioned that pacing should not equate to stopping progress, and warned of risks such as losing control of the future to AI or concentrating too much power in a few actors.

Google DeepMind co-founder and chair Demis Hassabis supported Amodei’s direction, calling it the “right path forward,” though he noted that details still need to be worked out. Hassabis also referenced a framework for an AI standards body he published in July.

Donald Trump criticized the idea of slowing AI development in a Truth Social post, asserting that strong presidential leadership and existing regulatory powers are sufficient to control AI risks. He framed calls for a slowdown as part of a “sick conspiracy” against AI and data centers, and argued that such efforts primarily benefit China.

U.S. Vice President JD Vance questioned the motives of AI companies advocating for regulation, calling it “a little bit weird” that frontier AI firms are “begging the government to regulate them.” He suggested this could be a “trojan horse” and reiterated concerns about China’s long-term AI ambitions.

House Speaker Mike Johnson told CNN that he supports convening a meeting with AI leaders and policymakers to determine the right balance between safety and national security, warning that a moratorium could allow China to overtake the U.S. Johnson said he would organize such a meeting “tomorrow,” though no timeline was confirmed.

Former Transportation Secretary Pete Buttigieg criticized inaction by Congress and the White House on AI safety, calling it “deeply dangerous.” His remarks were made in response to a Wall Street Journal column by Peggy Noonan that called for pausing AI development for humanity’s sake.