After White House artificial intelligence adviser David Sacks got President Trump to scrap a planned executive order in May that would subject AI models to an extended government review, senior administration officials including Scott Bessent and Susie Wiles were caught off guard and unhappy.
Wiles, the White House chief of staff, told others that Sacks had persuaded him to do so in a last-minute call, people familiar with the matter said. Treasury Secretary Bessent worried that AI-led cyberattacks could wreak havoc on the nation's banking system, the people said. A White House team had spent months negotiating to increase oversight of AI and had consulted a range of industry executives on the order.
At the urging of Wiles and Bessent, Trump later signed a slimmed-down version of the order.
Even as Trump dismisses calls for tighter federal regulation of the industry, the quiet freakout among some of his most senior aides over what could be one of the administration's most consequential policy decisions has begun.
In the months since the dispute, Wiles, Bessent, and other officials have gathered almost daily to discuss risks in the sector and consider potential guardrails, the people said, while trying to persuade Trump to take a more hands-on approach. White House officials have regularly held calls with executives including OpenAI's Sam Altman and executives at Anthropic to discuss concerns.
The divide among Trump's AI advisers-both inside and outside the government-continued in the weeks that followed, with Wiles, Bessent and National Cyber Director Sean Cairncross often pushing for more government scrutiny. Sacks, a venture capitalist, Meta Platforms Chief Executive Mark Zuckerberg and Nvidia's Jensen Huang are among those frequently urging the president to continue taking a light touch-an approach that appears to be winning the day.
Zuckerberg, Huang and SpaceX CEO Elon Musk recently spoke to Trump about their concerns with one plan for an industry-funded regulator, and successfully stalled the plan, people familiar with the matter said.
Sacks has become a particular point of friction for some White House officials.
He advocated for limited government involvement in AI in his initial role as the administration's AI and crypto czar. Some administration officials privately researched Sacks's businesses to see if the venture investor could personally profit by blocking regulations.
Sacks held his initial role as a special government employee, which meant the Office of Government Ethics required his venture firm to divest some, but not all, of its AI holdings. His firm's multibillion-dollar portfolio includes stakes in Elon Musk's SpaceX and many AI startups. Sacks previously invested in Zuckerberg's Meta, as well as Palantir and Airbnb.
He left his czar role in March to be co-chair of a White House tech advisory council. In the past six months, other administration officials have grown concerned about how far the technology has come, and how little the administration initially did to scrutinize it, people familiar with the matter said.
Sacks's power was on display again this week, when Trump declared Monday that the only AI regulation the U.S. needs is a "strong and smart president," following a call from lawmakers and some AI CEOs for a coordinated slowdown of model development to prevent cyberattacks or other disasters.
Hours later, Trump called in to an "All-In" event hosted by the tech podcast that Sacks co-hosts. Sacks was on stage with Huang at a live taping, when Huang, clad in his usual black leather jacket, took the call.
Huang praised Trump's earlier comments, then put him on speaker. "The robots will not be taking over. The AI will not be taking over the rest of the world," Trump said.
White House spokeswoman Liz Huston said: "The Trump administration is committed to unleashing American innovation, supporting the safe and successful deployment of the world's most advanced AI models, and protecting our national security."
A 'red flashing light'
The preview and gradual release of Anthropic's powerful Mythos model in the spring sparked part of the shift within the White House, after officials realized that the AI tool could be used to launch such attacks and pose national-security threats. Inside the administration, the Mythos model was a "red flashing light," a senior administration official said, prompting round-the-clock calls for days.
In July, OpenAI's hack of the company Hugging Face involving hundreds of AI agents acting in coordination fueled fears among officials of out-of-control models.
White House officials have since regularly battled with other AI executives, who have opposed many regulatory efforts. At times, the executives have gone directly to Trump, the people said.
Some of the White House aides have had to learn about AI as they go along and have quizzed those in the industry about how certain things work, people familiar with the matter said.
The White House team has begun looking at how rogue developers or foreign nations could use the technology in ways that hurt people and U.S. infrastructure such as local water and power utilities, a concern for officials including Vice President JD Vance. Some of the evaluations have included the National Security Agency and the Central Intelligence Agency, the people said.
As White House discussions about AI have expanded to how it could be used to create biological weapons, the meetings have at times included top HHS officials, they said.
Despite the concerns from his top advisers and much of the public, Trump has remained resolute that the nation needs to accelerate AI development and continue to build the data centers that are powering the AI boom, to stay ahead of China and boost the economy. "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!" he wrote on Truth Social on Monday.
Trump has defended the data centers despite polling showing most Americans oppose them in their communities, and warnings from his outside advisers that his stance could make him seem out of touch with the public. Sacks said Wednesday at a Politico event that existing product-liability laws and measures preventing fraud are largely sufficient regulations for AI models at this time.
"It is the responsibility of these companies to maintain control of their products," he said.
Some administration officials have told executives at AI companies that it is difficult for the White House to build consensus when the sector is often divided and CEOs contact Trump directly, people familiar with the matter said.
Trump's own aides have had little luck convincing him that AI has significant downsides, people familiar with the matter said. The president is "interested" in AI but "doesn't use it all the time," a senior administration official said.
Tech and AI companies have been prodigious donors to Trump and his affiliated organizations in this term. OpenAI's Altman has repeatedly attended fundraisers with Trump. Another top executive at the company, Greg Brockman, and his wife, Anna Brockman, have given $25 million. Nvidia's Huang has given money to Trump's new White House ballroom. Meta and other tech giants have made similar pledges.
Last month, some industry executives worked again to persuade Trump to scuttle another potential order.
Demis Hassabis, chief scientist of Google parent Alphabet, had been pitching White House officials on an industry-funded group that would set standards and prevent harm, similar to the Financial Industry Regulatory Authority, which oversees brokerage firms, in recent weeks. Many industry executives supported the concept, but Zuckerberg, Huang and Musk didn't, expressing concerns about who would be chosen for the body and the potential to concentrate power in OpenAI, Anthropic and Google, according to people familiar with the matter. The executives told Trump in separate conversations in recent weeks about their concerns and preference for the administration's current light-touch approach, the people said.
Trump ultimately didn't agree to setting up such a group, the people said, frustrating some inside the White House.
The start-and-stop policy efforts have confused industry executives and observers who say the direction of AI policy is still changing frequently nearly two years into Trump's second term. In turn, White House officials say they have been frustrated by mixed messages from AI executives, including the latest calls to coordinate a slowdown in model development with the government.
Vance told reporters this week: "I feel a little bit weird about the fact that you have so many frontier AI tech companies kind of coming to the government and begging the government to regulate them. It feels a little bit to me like a bit of a Trojan horse." He said the administration has some concerns but wants to regulate smartly.
Trump is expected to host AI CEOs at the White House next week on the topic, in a meeting set up with House Speaker Mike Johnson. Huang is among the CEOs expected to attend a state dinner with China's leader Xi Jinping.
"There is a wide aperture of possibility," said Hamza Chaudhry, head of national security policy at the Future of Life Institute, a nonprofit focused on reducing catastrophic risk from AI, "and also an administration still trying to figure out where it will land."
Comments