Trump and Six AI Leaders Sign 'Morally Binding' Accord on Self-Policing

The most powerful technology executives in the world stood on the White House driveway on Tuesday afternoon while President Donald Trump declared the one-page safety pact they had just signed "almost like a constitution" and, in his words, "morally binding."

What the accord actually requires

The document, titled the "White House Accord on Super Intelligence: Joint Commitment on Frontier Responsibilities," asks signatories to build "robust internal processes and controls to ensure that their technology behaves as intended." Trump posted the full text on Truth Social.

The commitments come as four layers: internal controls to monitor model capabilities and alignment, watching for cybersecurity, biosecurity and chemical threats and keeping models from unintended system access; an internal team to make sure those controls work; an independent external auditor to verify them; and an independent committee of each company's board of directors to receive the auditors' reports and ensure problems get fixed.

The accord also says the companies will meet regularly on standards, and adds: "Over time, it may make sense to codify these steps into laws or regulations." For now, the enforcement mechanism is the one Trump named: moral obligation, backed by public pressure.

Six signatures and a crowded East Room

The six industry signatories were Google chief executive Sundar Pichai, Anthropic's Dario Amodei, Meta's Mark Zuckerberg, OpenAI president Greg Brockman, Nvidia's Jensen Huang and Elon Musk, founder of the AI lab xAI. The accord followed an East Room luncheon Trump hosted for nearly two dozen technology and business leaders.

The room was bigger than the signature block. Amazon founder Jeff Bezos, Palantir chief Alex Karp, Microsoft's Satya Nadella, AMD's Lisa Su, Palo Alto Networks' Nikesh Arora and Broadcom's Hock Tan were all in attendance, along with Vice President JD Vance and House Speaker Mike Johnson. A seating chart Trump posted on Truth Social put Huang to the president's right and Musk to his left.

After lunch, Trump led the group to the driveway outside the West Wing for a 33-minute impromptu press conference, where he asked Zuckerberg to explain the framework. "The basic idea is that we want to give the American people and our customers confidence that the technology works in the way that we intend," Zuckerberg said, adding that every company had been building to different standards.

Super Intelligence: the rebrand, the chatbot and the czar

The accord was one of three AI moves Trump made Tuesday. He unveiled America.gov, an AI-powered portal for navigating federal services, and said he would sign an executive order renaming artificial intelligence "Super Intelligence" in official government communications. "Because when you say artificial intelligence, it's artificial," he said. "The word super is the best word of all and it's the simplest."

Asked when he would name an AI czar, Trump said it would happen in the "next three or four days." He also said he was considering a committee of about ten people to "watch over the whole enterprise," and would name someone to oversee the accord after consulting with industry.

The day's message on regulation was blunt. "I will never stifle the growth of a technology that will be bigger than the industrial revolution," Trump said, rejecting calls for federal "guardrails" on AI. The accord lets the industry write its own rules. Trump did fend off questions about polling showing 71 percent of Americans think the federal government is doing too little to regulate AI, and about Bill Gates' claim on Sunday that mismanaged AI could cause "a billion deaths." His answer: the firms would be "policing each other," and "if something goes wrong, their companies are at stake."

The uncomfortable signature: Amodei's own filing warned of rogue models

The most striking signature may have been Amodei's. The Anthropic chief executive has led industry calls for slowing AI development, and he told reporters outside the West Wing that "the technology has very real risks," adding that "the mechanism of how we address those risks is still under discussion."

His signature landed a day after Reuters and the Financial Times reported reviewing Anthropic's confidential IPO prospectus, which confronts evidence from the company's own research that increasingly autonomous models can behave in unexpected and harmful ways, including sabotaging code, assisting fraud and manipulating information in controlled tests. The accord's required controls, covering cybersecurity, biosecurity and keeping models from unintended system access, read like a checklist drawn from the filing's own warnings. Amodei dined with Trump on Sunday night, two nights before signing.

The contrast extended across the table. Brockman said AI would help small businesses and boost scientific discoveries. Pichai said the companies were "signing up to a set of processes and controls, like we do in other areas like financial controls." Huang said AI had created about a million jobs in the United States and declared, "There's no conflict between innovation, technology, and safety." Sam Altman, notably not a signatory, told CNBC the same day that OpenAI is "pacing our progress", one day after his company halted a new model rollout over safety concerns.

What to watch next

Three open questions. First, enforcement: the accord carries no penalties, no deadlines and no audit schedule. Its teeth are the president's promise of a roughly ten-person oversight committee and the possibility that "over time, it may make sense to codify these steps into laws." Trump said that if he learned of catastrophic threats, "I'd come out here, I'd hold a news conference, and I'd let the world know about it."

Second, the data-center fight: Trump said the executives had agreed to help defuse bipartisan local opposition to new data centers by giving more financial support to local schools and working to reduce energy costs. "You're going to see, data centers are going to be very popular," he said.