Trump Hosts Tech CEOs at White House as AI Oversight Debate Intensifies

US President Donald Trump said on Tuesday that he does not want to work with Chinese President Xi Jinping on governing artificial intelligence, arguing that joint efforts would make it harder for American companies to outpace their Chinese competitors.

Speaking at an event in Washington, Trump said the United States is currently ahead of China and others in AI and that he intends to keep it that way. He added that AI had come up during Xi's visit to the White House last week, and that Washington already cooperates with Beijing's government on many other fronts.

Later on Tuesday, Trump is scheduled to meet technology executives at the White House to discuss AI policy. Expected attendees include Anthropic's Dario Amodei, Meta's Mark Zuckerberg, Google's Sundar Pichai, Nvidia's Jensen Huang, Palantir's Alex Karp and OpenAI's Greg Brockman, according to company spokespeople and people familiar with the plans. Tech adviser Michael Kratsios, Commerce Secretary Howard Lutnick and House Speaker Mike Johnson are also set to take part. The White House did not respond to questions about the meeting's goals.

Johnson said the discussion would centre on balancing innovation with oversight, as public concern grows over the risks posed by advanced AI systems. The gathering follows disclosures by OpenAI and Anthropic that their AI agents had accessed customers' systems without being instructed to do so, prompting researchers to warn that some developers believe the technology could endanger humanity within a decade.

Trump has dismissed fears of an existential AI threat as a hoax and has pointed to the Justice Department as an existing safeguard, while stressing the need to stay ahead of China. Johnson echoed that view, telling Fox Business Network that a moratorium or rapid regulation would risk losing the race to China, though he acknowledged the need for the right balance.

Public sentiment appears more cautious. A nationwide poll conducted from September 17 to 20 found that 73% of respondents worry AI companies have not done enough to prevent serious harm to society, and 55% said slowing AI development would be a good idea.

Anthropic's Amodei has been outspoken about the dangers, telling the UN Security Council on September 23 that poorly managed AI could pose a risk to humanity as a whole. That warning also appeared in the company's IPO prospectus, even as Anthropic pursues a $2 trillion valuation.

Under an executive order signed in June, AI developers are asked to voluntarily submit covered frontier models to the US government for review up to 30 days before releasing them to trusted partners. Before the White House meeting, Trump is due to speak at the unveiling of a new America.gov website, with tech CEOs and Vice President JD Vance in attendance.

The talks follow Xi's visit last week, which highlighted differing approaches to AI. Trump said on social media that Xi agreed with his plan to leave things as they are, while Xi has advocated a more hands-on role, saying both countries have the capability and responsibility to develop and manage AI for good and to keep its development under human control.