President Trump secured a voluntary “constitution” from top AI executives at the White House, calling the self-policing pact morally binding while rejecting heavy-handed rules that would cede the edge to China.
President Donald Trump told reporters Tuesday that leading artificial intelligence executives had signed what he described as a “constitution” to police their own industry after a White House luncheon, framing the deal as protection without the red tape that could slow American innovation.
ABC News reported that the document is titled “The White House Accord on Superintelligence: A Joint Commitment on Frontier SI Responsibilities,” and that Trump cast the industry as “self-policing” with an agreement that is “morally” binding.
House Speaker Mike Johnson, who attended, stressed the pact is voluntary. He said it sets expectations for robust internal controls and layers of internal and external review, while Congress would keep guiding safe development without slamming the brakes.
The guest list was a who’s who of frontier tech: Meta CEO Mark Zuckerberg, Elon Musk, Anthropic CEO Dario Amodei, Google CEO Sundar Pichai, Nvidia’s Jensen Huang, Palantir’s Alex Karp and Shyam Sankar, Amazon’s Jeff Bezos, and OpenAI’s Greg Brockman, along with members of Trump’s Cabinet.
The meeting took place Tuesday at the White House, with an East Room luncheon followed by driveway and West Wing remarks. Trump later shared a screenshot of the accord; the full official copy was still being sought by reporters that evening.
Breitbart reported that Trump and the tech leaders signed the White House Accord on Super Intelligence as a voluntary joint commitment on frontier responsibilities, emphasizing internal controls and external review. Trump also signed an executive order rebranding artificial intelligence as Super Intelligence.
Trump put the stakes in plain terms. He said the biggest names in the world signed the document, he signed it as president, and it amounts to a form of protection. He called it almost like a constitution.
That stance fits a broader pattern of Trump pressing hard on security and national interest without letting ideology choke growth, much as he has taken firm public lines on other flashpoints from Iran diplomacy to domestic institutional overhauls.
The New York Post reported that Trump released the text of the voluntary accord and that it lays out four layers of internal controls, audits, and oversight for frontier AI models, covering cybersecurity, biosecurity, and related risks. Signatories included Zuckerberg, Musk, Amodei, Pichai, Huang, and Brockman.
Zuckerberg called the deal “a start and an accord that the whole industry could come to.” He described principles and commitments around building multiple layers of auditing and controls, with company boards independently reviewing auditor reports. “I think that these are good steps forward,” he said.
Pressed on why the White House did not demand more stringent guardrails than a voluntary document, Trump answered that the AI CEOs “love” America. Asked whether love is enough, he pointed to his head and said it requires that, and that they are the smartest people in the world.
He floated another idea: forming a committee of maybe 10 people, possibly drawn from the same group, to watch over the whole enterprise. Details on membership and authority were not locked down.
Reuters reported that after the meeting Trump announced the executives had agreed to voluntary AI standards, which he signed and cast as a major protective framework. Companies committed to work with independent auditors to assess systems and prevent unintended access, while Trump also pushed data-center expansion and said these powerful firms would make massive contributions to communities.
Johnson, speaking two weeks earlier about the planned session, said there would be a deliberate discussion of companies’ responsibility to maintain safety and what role, if any, government should play. On Tuesday he described the finished product as a statement of principles, standards, and voluntary commitments.
“It sets forth expectations. It’s in short summary robust internal controls and layers of internal and external review,” Johnson said. “That’s important because you need people to be providing oversight and transparency, and that’s what these leaders are going to do.”
He added that Congress would “continue to guide the industry’s development in a safe manner,” stopping short of any new enforcement hammer. The accord itself recognizes the steps are not yet written into formal laws or regulations, while leaving that door open later.
Johnson drew a hard line against a development freeze. “We cannot have a moratorium on the development of AI, because then we will lose our edge to China, and that has serious national security implications,” he said.
On Monday he went further on Fox Business, dismissing doom talk as political theater. “This is what Democrats do. They glom on to something and say, ‘This is doom and gloom and everyone will die if we don’t win,’” Johnson said. “Don’t buy it. It is a Chinese psyop.”
That national-security framing lands against a backdrop of sharp partisan reaction to Trump’s agenda, including polling that shows deep Democratic hostility to his return to power.
Just last week at the United Nations General Assembly, Trump said the United States “rejects” a global attempt to regulate AI. Several of the same tech figures had also attended a White House state dinner for Chinese President Xi Jinping. Amodei dined with Trump late Sunday night before the larger Tuesday session.
OpenAI and Anthropic had recently committed to third-party evaluators. The new accord, as described, leans on each company’s individual duty to implement controls and audits and to build trust with customers and the public.
Just the News reported that Trump planned to sign an executive order Tuesday evening officially renaming artificial intelligence as “Super Intelligence,” following the meeting with top AI leaders. He said it is not artificial, that everyone agreed on that, and that the companies came together out of love for the country and the world.
“They love our country and they love the world and they came together and they did something. It’s almost like a constitution in a way. It really is a form of protection... When they signed that, it meant a lot,” Trump said.
The same day carried more administration AI moves. RealClearPolitics reported that Trump signed the order directing the government to call AI “super intelligence,” that executives at the lunch signed onto the morally binding statement of principles with guardrails, and that the administration launched America.gov as an AI-powered front door to federal services while ordering agencies to integrate with it.
“This is one more example of why I will never strangle growth technology that will be bigger than the industrial revolution,” Trump said at the America.gov event. After the roughly two-hour lunch with about 20 top executives, he told reporters there was “a lot of commonality in this room, and all for the good.”
Trump’s preference for results over process theater also shows up in other corners of the executive branch, from personnel shifts inside federal law enforcement to hard calls on overseas security threats that leave little room for drift.
Critics on the left have already floated maximalist responses to Trump’s broader second-term project, including talk of fresh impeachment theories if Democrats regain the House. The AI session offered a different picture: industry titans in the room, a signed code of conduct, and a White House unwilling to hand Beijing a free pass.
Gaps remain. The full public text and complete signatory list were still thin on Tuesday night. Whether any future 10-person committee would carry real authority is unsettled. And lawmakers continue to weigh what, if any, statutory role government should take beyond the voluntary frame.
What is not unsettled is the direction. Trump is betting that American companies, bound by reputation, boards, auditors, and moral commitment, can keep the lead without a regulatory chokehold that freezes progress and invites a Chinese sprint past U.S. labs.
Self-policing by the people building the technology beats a freeze written in Washington or a rulebook written at the U.N. That is the bet the White House just put on paper, and the one Beijing would love to see America abandon.
