President Donald Trump said the United States will allow tech companies to “self-regulate” their artificial intelligence models, after convening a group of 31 industry leaders to sign a voluntary accord to audit AI development.
“We have a very big lead, and we’re going to keep our lead,” Trump told reporters outside the West Wing. “Our companies are dominating the world. They think that the federal government is doing enough, but they’re dominating China. They’re dominating everything. We want to keep it that way.”
The president pushed back against mounting concerns about the national security risks of AI following a recent incident that saw OpenAI agents unexpectedly access data from U.S. government websites.
Rather than pausing development and seeking federal regulations, the group of AI leaders signed what Trump called a “morally binding” accord focused on policing their own models in an effort to outpace China and other adversaries.
Trending
Meta Chief Executive Mark Zuckerberg said the agreement includes commitments around building “robust internal controls and detecting if there are any issues with the technology.”
The accord — the details of which were not immediately released — essentially means companies will voluntarily implement multiple layers of internal and external audits, which will be reviewed by each enterprise’s boards of directors, Zuckerberg said.
“We want to give the American people and our customers confidence that the technology works the way we intend,” he said as Trump and other AI leaders held court with reporters.
The White House held a luncheon earlier with prominent tech figures including Elon Musk, Nvidia’s Jensen Huang, Anthropic’s Dario Amodei, OpenAI’s Greg Brockman, Amazon’s Jeff Bezos, Palantir’s Alex Karp, Google’s Sundar Pichai and Microsoft’s Satya Nadella.
The accord coincided with OpenAI postponing the rollout of its GPT-6.1 Astra model until an extensive safety review can be conducted. Earlier this month, rogue agents accessed data from the Securities and Exchange Commission, Department of Education and Census Bureau without authorization — the second known incident of this nature in a matter of months.
OpenAI CEO Sam Altman told CNBC on Tuesday his company is deliberately pacing its progress in an effort to advance its safety goals.
Altman, Amodei and Musk have called on their peers to slow AI production until meaningful safeguards can be put in place to ensure such incidents don’t endanger Americans or national security. Amodei also held a one-on-one dinner with Trump on Sunday ahead of the broader summit.
Huang has said that the AI industry does not need new laws or regulations and that he believes existing legal frameworks and market forces are already sufficient to handle safety. Trump lauded the Nvidia executive Tuesday.
“We build artificial intelligence to benefit everyone,” Huang said. “We think that under President Trump’s leadership, everyone is coming together to agree on how to move forward safely.”
At a state visit last week, Trump and Chinese President Xi Jinping agreed to establish a communication channel for AI-related incidents, but both Trump and House Speaker Mike Johnson emphasized at the luncheon Tuesday that the administration prefers limited cooperation because the U.S. holds a definitive lead.
“We cannot lose the race to China,” Johnson said. “We cannot lose America’s edge because it’s important for every American worker, every American family, for their safety, security, the future, the economy. But it has to be a balance.”
The result of Tuesday’s summit drew consternation from Democrats, who have called for stricter regulations on AI companies.
Sen. Bernie Sanders, a Vermont independent who caucuses with the Democrats, has led the push to temporarily halt the development of advanced AI systems until Congress can advance federal safety guidelines.
“AI is planning to spend upwards of $300 million this election cycle,” he said on X Tuesday. “The result: Congress does not regulate AI. AI regulates Congress.”
Author
