Regulation · U.S.
Trump tells the UN that US documents will say 'super intelligence' and that the Justice Department will watch AI
President Trump told the 81st United Nations General Assembly on September 22, 2026 that United States documents will use the term "super intelligence" in place of "artificial intelligence," and that the administration will watch the sector through the Justice Department rather than new regulation. No executive order, memorandum or Federal Register document had made the change as of that evening.
President Trump told the 81st session of the United Nations General Assembly on September 22, 2026 that United States government documents will stop calling the technology artificial intelligence and start calling it super intelligence. The statement to the general debate is on the United Nations’ own record of the session. He also said the administration would police the industry through existing law enforcement rather than new rules.
What he said
As The Hill transcribed the passage, Trump said: “The use of the word artificial makes intelligence sound fake. It is not fake. It’s actually amazing. But we have to be careful. In fact, it is exactly the opposite of what it purports.” He continued: “From this point forward, all of United States documents and hopefully the world’s will be changed to use the much more accurate term super as opposed to artificial,” and then, “In other words, welcome to the new world of super intelligence, SI. Let’s see if that goes. It sounds much better. It is much better, and it’s much more accurate.”
Reuters reported the same announcement and two further remarks. On oversight, Trump said the administration would watch the sector “closely through the Department of Justice”. On regulation, he said: “I’m not going to stifle growth of something that will be bigger than the Industrial Revolution.” Reuters wrote that he has dismissed alarm about the technology as a “hoax” in recent weeks. PBS quoted him rejecting “any attempt to construct a globalist scheme to control artificial intelligence”.
Transcripts of the passage differ in small ways. The Hill and PBS render one clause differently, and neither is an official text. The United Nations video of the statement is the record of what was said.
Nothing has been issued
A change in what officials call the technology takes effect through written instruments, and none had appeared by the evening of September 22:
- The White House presidential actions page carried nothing later than September 18, so no executive order or memorandum directing the change had been published.
- A Federal Register search for “super intelligence” returns no documents at all, presidential or agency.
Until something is issued, the announcement binds no agency and changes no filing, rule or form.
What the law calls it
The term is defined by statute. 15 U.S.C. 9401(3) provides that “the term ‘artificial intelligence’ means a machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations or decisions influencing real or virtual environments.” That definition is carried into agency rules, grant conditions and procurement documents that cite it. A change in vocabulary at the podium does not amend it, and only Congress can.
The same is true of state law, which is where most of the duties on developers now sit. New York’s RAISE Act, article 44-B of the General Business Law, applies to a “large frontier developer” and a “frontier model” as section 1420 defines those terms, and its obligations begin on January 1, 2027 whatever Washington calls the technology. A state legislature is not bound by federal usage, and nothing in the announcement reaches a state statute.
For a lawyer, the practical question is narrow. If agencies do begin writing “super intelligence” in guidance, contracts and solicitations while statutes and regulations keep the defined term, documents on both sides of the same transaction will use different words for the same thing. That is a drafting problem, handled the way any other definitional mismatch is handled: define the term once in the agreement and say what it maps to.
The week around it
The remarks land in a week when the administration has been drawing a line between encouraging the industry and absorbing its risks. On September 21, Treasury Secretary Scott Bessent told CNBC that the July intrusion into Hugging Face’s systems was “the responsibility of the OpenAI management, not a bunch of agents”, and that the administration would not take liability off the labs’ hands (CNBC transcript). Read together, the two statements point the same way: no new regulatory regime, no liability shield, and enforcement through the agencies and courts that already exist.
What that leaves is the question of which law. The speech named no statute, no division of the department, no investigation and no enforcement program.
What is not yet known
Whether any written directive will follow, and what it would cover: agency guidance, procurement documents, regulations, or only speeches and press material. Whether the phrase would carry a definition of its own, and how it would sit with the statutory term, which the administration cannot change on its own. What the reference to the Justice Department means in practice, since the speech named no statute, no investigation and no enforcement program. The White House had not answered those questions in the sources reviewed.
Filed under
RegulationU.S.The story on LexCircuit