Trump Directs Federal Agencies to Adopt Super Intelligence Terminology
President Donald Trump signed an executive order on September 29, 2026, requiring U.S. executive-branch policy websites, policy documents and press releases to use 'Super Intelligence' instead of 'artificial intelligence'. The Verge The order is titled "Inaugurating the Era of Super Intelligence." The White House The White House lists 'Inaugurating The Era Of Super Intelligence' as an Executive Order dated September 29, 2026.
The directive instructs executive departments and agencies to use the terms "Super Intelligence" and "SI" in official correspondence and public communications, "to the maximum extent permitted by law." The White House It does not require agencies to change past regulations or documents. Under the order, officials are expected not to acknowledge the use of the terms 'artificial intelligence' or 'AI' in relevant settings.
Trump had first mentioned the plan to rebrand artificial intelligence as Super Intelligence in a speech to the United Nations General Assembly. He signed the order after hosting a White House luncheon with tech CEOs and government officials to discuss Super Intelligence. Nvidia CEO Jensen Huang and Tesla and SpaceX CEO Elon Musk appeared alongside Trump in public remarks after that meeting.
Trump had said he would sign an executive order renaming artificial intelligence to 'super intelligence'. CNN Top AI executives signed a commitment to 'self-police' AI development after meeting with Trump. Trump rejected new AI laws after the executives agreed to that approach. SMH Separately, the administration launched an America.gov AI website intended to make government services more accessible.
Looking at what this means for practitioners, terminology matters. Federal policy language feeds directly into procurement requirements, compliance filings, grant solicitations, standards references and contractor documentation. A mandated swap from AI to SI does not change model architectures, training pipelines, evaluation harnesses or deployment controls, but it does change the words that must appear on the cover page. Teams that track federal guidance will need translation layers. Search matters. Version control matters. Audit trails matter.
In this author's view, the friction will be practical rather than theoretical. Engineers already maintain mappings between internal system names and external regulatory terms. This adds another mapping. Worth flagging, the non-acknowledgment expectation for 'artificial intelligence' or 'AI' creates ambiguity for public communications staff and for vendors that must quote federal text verbatim while maintaining consistency with their own technical documentation, customer contracts and safety reports. That work is unglamorous. It still consumes time.
The broader context here is the parallel commitment to 'self-police' development and the decision not to pursue new AI laws. For technology leaders, voluntary commitments can move faster than rulemaking and can reflect operational knowledge from labs running large-scale training and inference. They also lack the enforcement mechanisms, reporting obligations and judicial review that come with statute. My own experience watching my children grow up with successive platform shifts suggests adoption rarely waits for vocabulary to settle. People learn the tool in front of them, whatever agencies call it.
The optimistic case is straightforward. Clearer citizen-facing services, if the America.gov AI effort delivers on accessibility, would be a tangible benefit. Consistent federal branding around advanced systems could focus attention and funding. The risk is confusion if one label is applied to everything from rules-based automation to frontier foundation models. Precision helps builders. For now, builders should treat SI as a required administrative term for federal-facing material, keep AI as the working technical term everywhere else, and document the equivalence explicitly.
