President Donald Trump stood before the United Nations General Assembly in New York and announced that the age of “artificial intelligence” is over, at least in name. “Welcome to the new world of super intelligence — SI,” he declared, and with those words he set off a semantic storm that reached far beyond the diplomatic hall. To Trump, the phrase “artificial intelligence” has a fundamental problem: it makes the technology sound fake. It sounds manufactured, cold, maybe even a little threatening. He wants something that captures the scale and promise of machines that write, paint, diagnose diseases, drive cars, and analyze vast oceans of data. So from this point forward, he announced, all United States documents — and, in his words, “hopefully the world’s” — would use the term “super intelligence” instead. The decision did not come out of nowhere. In the days before the speech, Trump had turned to his Truth Social platform and asked followers to help him choose a new name. The options were as bold as they were vague: superior intelligence, extreme intelligence, supreme intelligence. In the end, “super intelligence” won the online vote. Trump presents the rebrand as a matter of accuracy, but many people see it as a matter of image. This is not the first time he has tried to shape language from the top down. He has already ordered changes to maps and federal communications, renaming the Gulf of Mexico and even Lake Ontario. Those moves were steeped in political symbolism, and they were fiercely debated. But this latest rebrand targets something more universal than a patch of water. It seeks to redefine the vocabulary of the twenty-first century’s defining technology. For his supporters, it is a sign of ambition. For critics, it is a sign of hubris. The truth, as usual, lies somewhere in between — and it is not just about words. Language shapes how people think, what they trust, and how much power they are willing to hand over to the machines in their lives.
To understand why researchers are raising eyebrows, it helps to understand what “superintelligence” has traditionally meant. The term is not new. It carries a specific, carefully defined meaning in computer science and philosophy. More than a decade ago, the philosopher Nick Bostrom wrote a celebrated and unsettling book called Superintelligence. In it, he described an intellect that surpasses the brightest human minds not in one narrow area but in every domain: science, art, social wisdom, strategic planning, creative problem-solving. This is not just a smarter calculator. Bostrom’s superintelligence is a mind so far beyond our own that our attempts to predict its behavior are almost pointless. It may be a machine that improves itself in recursive loops, becoming increasingly powerful until it leaves human comprehension behind. And crucially, it is still hypothetical. No such system exists today. The most advanced AI programs are narrow. They can win chess games, write essays, generate realistic images, and summarize whole libraries of medical research. But they do not understand the world in any human sense. They do not have desires or goals of their own. They cannot decide what matters. They are astonishing tools, but they are still tools. Simon Coghlan, a digital ethics lecturer at the University of Melbourne, put it plainly: the rebranding is misleading. The word “superintelligence” suggests a self-improving mind, a system that learns and grows on its own with no reference to human design. Today’s AI does not do that. It depends on training data, human prompts, and limits set by its creators. To call it “super intelligent” is to give it credit for powers it has not yet earned. It is a little like calling a toddler who has learned to tie his shoes an Olympic athlete because he is fast. The ambition is understandable; the accuracy is not.
Experts across the world have been cautiously pushing back against the President’s vocabulary. Simon Coghlan, from the University of Melbourne, doubts the term “SI” will stick because it exaggerates what today’s AI can actually do. He reminded people that a truly superintelligent system, if it appeared, would be considered by many researchers to be the most consequential technology ever created. It would change everything, maybe even end the human era. That is not the world we live in today. Our current AI systems may surprise us, but they also make obvious mistakes. They hallucinate facts, reinforce biases, and stumble over basic common sense. Computer scientist Kokil Jaidka made a similar point in sharper language. Not all AI platforms today exude super intelligence, she said. The new name, in her view, borrows the prestige of a term researchers reserve for a much bigger leap. Ben Leong, a computer science lecturer at the National University of Singapore, agreed. He predicted that professionals would want to save “SI” for a higher level of AI, not use it to describe the products we have right now. Leong’s caution is telling. The people closest to this technology know how far it has come, and they also know how far it still has to go. They are used to the hype cycle. Every few years, some new phrase appears and gets overused, creating wild expectations that are impossible to meet. When the bubble bursts, the public grows cynical and the entire field pays the price. What may be a short-term public relations win for the White House could be a long-term credibility problem for AI research. The word “superintelligence” is not just a label; it is an aspiration. And aspirations should not be confused with achievements.
Inside the Trump administration, the new term was adopted with a speed that suggested careful choreography. “The Super Intelligence (SI) President has spoken! The United States is and will remain the global leader in SI,” wrote Ethan Klein, the U.S. Chief Technology Officer, on social media. Trump’s ambassador to the U.N., Michael Waltz, was just as enthusiastic. “If you didn’t see the president’s speech, it’s now super intelligence,” he said. “I think he’s absolutely right.” For the President’s inner circle, the rebrand serves a broader political purpose. It sounds victorious. It reinforces the idea that America is not just participating in the AI revolution but leading it. It also fits neatly into a style of communication that relies on repetition and slogan-like language. But the federal government’s ability to make the term stick is limited. Trump can order the State Department to use “SI” in cables and direct agencies to change their style guides, but he cannot dictate the language of Silicon Valley, academic journals, or the global media. When he said “hopefully the world’s” documents will follow, he was acknowledging as much. The U.S. government can name things. It cannot make names popular. That is a lesson that has played out many times in history. Official terminology often fails to catch on because language is a living, messy, democratic process. The public decides which words feel true and useful. Right now, “artificial intelligence” feels true to most people. It describes a thing that was created by humans, at artificial hands, and it acknowledges both the intelligence and its limits. “Superintelligence” feels like advertising: bold, uplifting, and untested. The Truth Social poll that produced the term was not a scientific process. It was a focus group for the faithful. And a focus group of believers is unlikely to change the way the rest of the world talks.
The international response was polite, skeptical, and in at least one case, slyly amusing. When Canada’s minister of AI, Evan Solomon, was asked for his thoughts on the new name, he answered: “Canadians are not enthusiastic, at all, about renaming things.” It was a direct jab at Trump’s habit of renaming landmarks, including a lake that exists to a large extent under Canadian jurisdiction. The exchange was a reminder that language in this administration is never neutral. Every name is a statement, and every statement is a move on a larger chessboard. Trump’s previous renaming decisions had already angered neighboring countries and historians. Renaming the Gulf of Mexico touched on deeply sensitive issues of identity and national heritage. If the President now presses the phrase “super intelligence” into official use, he is doing something even more consequential: claiming ownership of a global technology with a single American word. But the rest of the world does not have to accept it. Canada and other allies have been building their own AI policies, with a stronger emphasis on safety, transparency, and human rights. They have no interest in adopting a label that seems designed to market rather than describe. Solomon’s quip was therefore more than a wisecrack. It was a signal that the United States cannot simply re-brand a shared future by itself. In a world where AI research is increasingly international, a name that works in a campaign rally may not work in a treaty room. The technology does not recognize borders. Neither do its risks. If humanity is going to manage this transformation together, it will need a vocabulary that invites collaboration, not command.
At the end of the day, the fate of “super intelligence” will be decided by something older than executive orders and social media polls: adoption. Words survive only when people choose to use them. Talk to software engineers and they will keep saying AI. Talk to researchers and they will keep saying artificial intelligence. Talk to business executives and they will talk about generative AI, machine learning, and foundational models. The term “SI” may appear on some government memoranda. It may even dominate a few news cycles. But if the world’s actual AI community ignores it, it will dissolve into a historical footnote. That would be disappointing to its supporters, but not unusual. Language is full of failed rebrands. This does not mean the President’s proposal is entirely trivial. It reflects a real desire to change how society feels about technology. There is an argument to be made that “artificial intelligence” gives people the wrong impression. It suggests something fake or alien, when the reality is that these systems are embedded in our daily lives. Maybe a warmer, more confident name would help people engage more thoughtfully with the tools around them. But there is also a danger. Calling today’s AI “superintelligent” could inflate public expectations and create irrational trust. A person who believes a chatbot has superior intelligence might follow its advice without question. That is not a recipe for wisdom; it is a recipe for harm. The gulf between the name and the reality is where the trouble lives. Perhaps the term “superintelligence” belongs to a future generation, a time when machines truly think on their own. Until then, the humble phrase “artificial intelligence” is more honest. It acknowledges that this is human-made intelligence, with human limits, and that the story is still being written.