President Donald Trump publicly dismisses AI safety concerns as a "HOAX" — but behind closed doors, his own administration is quietly wrestling with a far messier and more urgent debate about whether the technology could destabilize the economy.
According to The New York Times, insiders privately say his combative public stance is really about market anxiety: he fears any pause in AI-driven growth could send the broader economy tumbling. Venture capitalist David Sacks has been especially influential in shaping that thinking, having successfully lobbied earlier this year to gut a planned executive order that would have forced new AI systems through government safety checks.
Meta's Mark Zuckerberg and Nvidia's Jensen Huang echo similar warnings to Trump directly, arguing that losing ground to China matters more than any domestic safeguards.
Inside the West Wing, though, that confidence has cracks. Unexpected AI failures over the summer pushed some administration officials who once championed unregulated growth to quietly acknowledge that scenarios they used to laugh off, like AI-enabled cyberattacks, dangerous biological applications, or interference with sensitive military systems, no longer feel so far-fetched.
Complicating matters further, the administration lacks any single official actually in charge of coordinating AI policy. National security adviser Marco Rubio rarely factors into the discussions, and the cyber-and-emerging-tech advisory role that existed under Biden simply no longer exists in this White House. The national cyber director role currently sits with a former RNC official with limited background in the subject. Staff turnover has compounded the confusion, with several White House officials working on AI security departing in recent months.
Anthropic CEO Dario Amodei, meanwhile, appears to have fallen out of favor entirely after publicly floating a plan last weekend for the industry to slow down and welcome outside safety inspectors, a pitch Trump rejected while accusing unnamed critics of politicizing the issue to undermine American competitiveness. By contrast, OpenAI's Sam Altman and Elon Musk, who've both offered general support for elements of Amodei's slowdown proposal, remain in regular contact with the president.
This comes as Anthropic whistleblowers express fears that AI technology could doom humanity if left unchecked.
Underlying all of it, insiders say, is Trump's belief that AI-driven growth represents his best shot at the manufacturing renaissance he's promised since his first term, when pledges to revive coal jobs largely fell flat. Having floated the idea of a Warp Speed-style federal push for data center construction as far back as a 2024 meeting with OpenAI executives, Trump seems unwilling to let anything slow the momentum now, even as chief of staff Susie Wiles and Bessent quietly organize internal discussions grappling with the technology's darker possibilities.
"The robots will not be taking over," Trump insisted this week, phoning into a tech conference. "The A.I. will not be taking over the rest of the world."