Policy & Regulation

Trump Unveils 'Morally Binding' AI Pact Signed by Top Tech CEOs

Trump announced a 'morally binding' AI pact, the 'Joint Commitment On Frontier Responsibilities,' signed by top tech CEOs, plus an order renaming AI 'superintelligence.'

By Marcus Bennett3 min read

Updated

Why it matters

  • Trump announced the deal on Tuesday, following a White House luncheon with top tech executives.
  • The pact is named the 'Joint Commitment On Frontier Responsibilities' and is described as 'morally binding.'
  • Trump said: 'It's almost like a constitution in a way, and the biggest people in the world signed that, and I signed it as president.'
  • An executive order officially rebrands 'artificial intelligence' as 'superintelligence.'
  • Trump characterized the arrangement as enabling 'tremendous self-policing' by the companies.

President Donald Trump announced on Tuesday that the heads of the largest US tech and AI companies have signed a "morally binding" agreement to place controls on artificial intelligence. The deal, called the "Joint Commitment On Frontier Responsibilities," emerged from a White House luncheon with top tech executives and was announced alongside an executive order officially rebranding "artificial intelligence" as "superintelligence."

Trump framed the agreement as a foundational document for the industry. "It's almost like a constitution in a way, and the biggest people in the world signed that, and I signed it as president, and it really is a form of protection," Trump said.

He also characterized the arrangement as enabling "tremendous self-policing" by the companies themselves — a phrase that captures the central tension of the announcement: the White House is presenting an industry-managed framework rather than a regulatory regime with enforcement powers.

What did the president actually announce?

Two things happened at the Tuesday luncheon, according to the announcement:

  • The "Joint Commitment On Frontier Responsibilities," described by Trump as "morally binding," signed by the heads of the largest US tech and AI companies and by the president himself.
  • An executive order officially rebranding the term "artificial intelligence" as "superintelligence" across government language.

The president's own description of the agreement is notably modest on enforceability. A "morally binding" commitment is, by definition, one without legal force. Trump's comparison to a constitution — a document that carries the weight of law — sits uneasily beside the voluntary nature of the signing, and the White House did not, in the announcement, detail specific obligations, verification mechanisms, or penalties attached to the commitment.

Why does the wording matter?

The stakes here are semantic as well as substantive. The rebranding of "artificial intelligence" to "superintelligence" via executive order changes the vocabulary of US government policy for the most heavily funded technology sector in the American economy. Superintelligence, in AI research, refers to a hypothetical system surpassing human capability across all domains — a far stronger claim than current frontier models can support.

Adopting that term officially conflates shipping products with speculative future capability. The result is a policy language that inflates what today's systems do while the actual governance instrument — the Joint Commitment — remains, by the president's own account, morally rather than legally binding.

What does 'tremendous self-policing' mean in practice?

The announcement offers little detail. What is known comes from Trump's own statements: the signatories are "the biggest people in the world," the president signed alongside them, and he considers the document "a form of protection."

The phrase "tremendous self-policing" signals that the administration's chosen instrument is industry self-regulation. That approach places responsibility for frontier AI controls with the same companies building frontier AI systems. Critics of voluntary frameworks typically argue they lack teeth; supporters argue they allow faster adaptation than statute. The announcement itself does not resolve that debate — it embodies one side of it.

Who signed, and what happens next?

The White House luncheon brought together top tech executives, and Trump said the heads of the largest US tech and AI companies signed on. The announcement did not enumerate the individual signatories or their companies in the details released.

The open questions after Tuesday are concrete ones: whether the Joint Commitment On Frontier Responsibilities will be published in full, whether it specifies any measurable obligations for frontier model development, and whether "morally binding" language will satisfy lawmakers and foreign regulators watching how Washington governs its most valuable industry. Until the text of the commitment surfaces, the enforceability of the pact rests entirely on the willingness of the signatories — and the president's word that it constitutes "a form of protection."

Source: The Guardian AI

Share this article:

More from Marcus Bennett

Marcus Bennett

Show full bio

Senior reporter covering consumer brands and retail at AI In Context.

170 articles

Related articles

  1. Trump Renames AI 'Super Intelligence' as CEOs Sign Non-Binding Pact
  2. Trump's White House AI Accord Leaves Safety to Companies Themselves
  3. Trump's AI Safety Deal: Tech CEOs Sign Self-Regulation Accord
  4. Trump Declares AI Renamed 'Super Intelligence' as Poll Numbers Sink
  5. White House AI Accord Misspells "United States" Under Trump's Signature

« Previous articleNext article »