President Donald Trump called for a ban on the term “artificial intelligence,” or “AI,” on Tuesday, writing in a Truth Social post, “In short order, we will be officially BANNING the term, ‘AI/Artificial Intelligence,’ in all relevant documents, and everything else, due to its inaccuracy.”
On September 29, Trump signed an executive order that officially replaced the term AI with “super intelligence.” EO 14434 directed federal departments and agencies to use the latter term or “SI” in all public communications, official policies, websites, and non-statutory documents, and to stop using “AI.”
Trump ordered the change, arguing that “artificial” makes the technology sound fake and possibly untrustworthy, and that “super intelligence” better reflects its real-world impact and capabilities.
“The terminology used by the Federal Government should reflect the transformative capabilities of these technologies and the limitless opportunities they create for the American people. Accordingly, the term ‘Super Intelligence’ more appropriately captures the promise, potential, and rapidly advancing capabilities of these technologies,” the White House explained.
Transition Underway
Federal agencies may begin to use the SI terminology in documents, including solicitations, reports, and policy statements. The National Institute of Standards and Technology (NIST), the agency that develops technology standards, measurements, and guidelines to support American innovation and economic security – and develops the voluntary AI standards, guidelines, and best practices – announced earlier this month that it would incorporate the term “super intelligence” or “SI” as directed.
The United States Navy issued a mandate to its commands to transition to the SI designation.
Several tech leaders, including Elon Musk, Jensen Huang of Nvidia, and Jeff Bezos, have publicly embraced or used the terminology. Musk has led the efforts, renaming some of his initiatives.
However, California Governor Gavin Newsom issued a competing executive order instructing state agencies to continue using the term “artificial intelligence.”
There Has Already Been Super Intelligence
Before the White House called for the rebranding of AI to SI, there already was something described as “Artificial Superintelligence,” or ASI, a hypothetical form of what we have traditionally called “AI” that is smarter than the best human brains in every field, including scientific creativity, general wisdom, social skills, and problem-solving.
However, Dr. Jim Purtilo, associate professor of computer science at the University of Maryland, questioned whether ASI was ever really an accurate term.
“‘Super intelligence’ is just a manufactured term in search of a substantive definition, which is to say, it is hype,” Purtilo told ClearanceJobs. “There is no point where ordinary AI suddenly becomes more powerful and crosses a threshold to being ‘super’ intelligence. Also, nobody is yet telling us the difference from ‘artificial generalized intelligence,’ another term tossed around, which mostly refers to AI exhibiting the ability to learn and function across a wider array of activities that are typically performed by humans.”
At the same time, the computer science professor also questioned whether SI is an accurate way to describe AI in any form.
He noted that human intelligence is far more nuanced and complex than the ability to solve puzzles quickly or accurately remember many useful details, tasks artificial intelligence tools can perform well.
“In fact, AI behaviors reflect a mathematical averaging of answers from many, many users on the Internet. Ask a question of all those people, and there may be outliers. Still, most will likely give a plausible answer, and the math behind large language models will construct a ‘best fit’ accordingly,” Purtilo continued. “Once you see what goes on behind the curtain, you can applaud the mathematics and effects, but any sense of something deeper going on will be gone forever.”
By extension, the same is true for superintelligence, which may actually make what we have been calling AI suddenly sound more capable than it is.
“You might be factoring in more parameters and solving bigger equations, but at the end of the proverbial day you’re just projecting trust in what are otherwise a bunch of sterile numbers. The term is great if you are a huckster trying to sell us those equation solvers all over again, but all they are doing is slapping a new label on the old bottle of magic elixir,” warned Putilo.
Changing a Long-Used Term
AI is commonly used today, so much so that it seems to have been around forever; yet, it was only coined in 1955 by computer scientist and mathematician John McCarthy, who first used the phrase in his official workshop proposal, titled “A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence,” which he submitted alongside Marvin Minsky, Nathaniel Rochester, and Claude Shannon.
Still, for decades, AI was a concept few heard of.
The term occasionally appeared in popular culture. Yet it wasn’t until IBM’s Deep Blue computer defeated chess grandmaster Garry Kasparov in 1997, or IBM’s Watson won Jeopardy! in 2011, that AI reached the mainstream. For most Americans, it wasn’t until the release of GPT-3 in 2020, which demonstrated large language model (LLM) capabilities, and the public debut of ChatGPT in late 2022 that conversational AI truly reached everyday consumers en masse.
Because AI is fairly “new” to most people, the term could catch on among average Americans, but other nations may not be on board. Some nations use alternatives, such as Germany and Norway, which have “KI” for “Künstliche Intelligenz.”
Other nations and US partners, like France, Spain, and Italy, have IA, respectively, for “Intelligence Artificielle,” “Inteligencia Artificial,” and “Intelligenza Artificiale.” Others have simply adopted the English AI rather than a local acronym. Will those nations change to SI?
Poland won’t have to, as it already has “Sztuczna Inteligencja.”
Then there is the fact that business adoption of AI across OECD nations continues to rise rapidly, reaching over 20% of firms overall and over 57% in information and communication technology (ICT) sectors. The .ai Internet domain has increasingly become globally synonymous with artificial intelligence, despite it being the country code top-level domain officially belonging to Anguilla, the small British Overseas Territory in the Caribbean.
The domain .si could be a bigger issue, as it is the country code top-level domain for Slovenia. Since the Trump administration’s EO, there has been a massive global registration surge from international tech companies and AI startups trying to register websites with the .si domain!
A New Metric or the US as an Outlier
Even if the U.S. fully embraces SI instead of AI, it may not set a new metric, but it could make the United States an outlier, much like how it doesn’t use the metric system. The White House sought to press foreign acceptance of SI over AI, but so far no other countries have officially changed their national laws, style guides, or standard government vocabulary.
Moreover, most major international media, tech institutions, and global regulators will likely continue to use the standard term “artificial intelligence,” viewing the U.S. rebrand as a domestic political and semantic shift rather than an international standard.
As with the Trump administration’s orders to rename the Gulf of Mexico the Gulf of America, Lake Ontario Lake America, and even to change the Department of Defense to the Department of War, a future administration could just as easily undo them.
Even if Congress adopts a new statutory definition of SI, that change could present implementation issues, noted Wiley Rein LLP, the D.C. law firm that focuses on regulatory compliance, international trade, government contracts, telecommunications, media, technology (TMT), cybersecurity, intellectual property, and election law.
“For example, agency materials have leveraged the definition of ‘artificial intelligence’ in 15 U.S.C. § 9401(3) in contract requirements, reporting obligations, grant programs, and agency guidance that are based on the current definition of AI, and many procurement-related policies, guidance documents, and Executive Branch requirements either expressly incorporate that definition or use definitions derived from it.”
Most importantly, the name change doesn’t really make the “intelligence” any smarter.
“A rose is still a rose no matter how you call it,” quipped technology industry analyst Roger Entner of Recon Analytics.
