Seventy years ago last summer, a small group of scientists gathered at Dartmouth College to kick around a wild idea: that machines might one day think, reason, recognize patterns, and even speak the way humans do. There was no roadmap and no clear funding; the whole thing felt less like a scientific program than an intellectual stunt, a bunch of brilliant people asking a question that had no known answer. One of the scientists, John McCarthy, suggested a name for this newborn pursuit: artificial intelligence. It was never a perfect name, and what does “artificial” intelligence even mean, after all? But it stuck. It has since become one of the most recognizable technical terms on Earth, perhaps the most recognizable of all. Search Google Scholar for “artificial intelligence” and you will get roughly 7,420,000 results. Governments put the phrase in policy papers, software companies put it in every product description, journalists use it as shorthand for anything with a processor and a pulse, and ordinary people have learned to be suspicious of it. Even those who most despise AI usually spend their energy worrying about what it does, not what it is called. The term is not merely useful; it is the air we breathe. Earlier this year, however, it turned out that one extremely important person never really liked the name. President Donald J. Trump has a specialty in renaming things that irritate him—gulfs, lakes, centers for the arts—and in late September, speaking at the United Nations, he floated the idea that “artificial intelligence” was an ugly, unnatural phrase. It sounded like rhetorical bluster. It wasn’t. This week, after a White House lunch attended by the most powerful AI executives on the planet, Trump signed an executive order banishing the phrase from official government usage. Federal officials are henceforth instructed to say “Super Intelligence.” So much for the name that survived seven decades of revolution.
To their shame, the lords of artificial intelligence—or should I say the lords of Super Intelligence—meekly went along. Not one of them stood up and pointed out that you cannot rename a field with a memo. They sat in the White House, presumably surrounded by glittering plates and silent minds, and allowed the most consequential technology of their lifetime to be rebranded by a man who has never once written a line of code. It was the ultimate presidential flex. There they were: executives worth trillions of dollars, men who have built systems that can write poetry, diagnose illness, drive cars, negotiate legal disputes, and maybe one day surpass the very people who invented them. In the average boardroom, any one of them is the alpha. But in that room, the alpha was the guy with the orange skin and the borrowed black suit. They may have had the majority of the money and the majority of the world’s data, but he had the only vote that counted. For them, this was supposed to be their foundation. Artificial intelligence wasn’t just a product line or a research division; it was their identity. They grew up reading science fiction under the covers, flashlight in hand, dreaming about thinking machines. The phrase “artificial intelligence” ignited something in them that has never really gone out. They built their companies around that dream, renamed themselves to reflect it, raised capital on the strength of it, and have spent the last four years talking about almost nothing else. And yet, when the president treated them to lunch and then decided to rewrite history in real time, no one made a sound. To object would have been easy enough. They are not government employees. They could have left, or issued a polite joint statement, or simply continued to use the old term and let the order die from neglect. Instead, they accepted the new vocabulary as if it had always been on the menu.
What is most damning is the total silence. You would expect at least one person in that room to have a flash of courage: Jeff Bezos, who has been pushed around by Washington before but also has the resources and the distance to say no; Jensen Huang, whose company has become the most important hardware maker in the AI boom; Dario Amodei, who spends his public life warning about catastrophic risk; Mark Zuckerberg, who has spent years bending to political winds; Sundar Pichai, whose company basically wrote the book on modern AI; Greg Brockman and Alex Karp, both eccentric enough to be contrarian; or Elon Musk, who has never met a fight he didn’t want to win. These are not shy people. They have survived legal battles, congressional hearings, public scandals, and the radical uncertainty of a world that might be transformed by their work. Yet in that room, they were all reduced to agreeable guests at a banquet. According to Jacob Weisberg, author of “Profiles in Cowardice: A Study of Collaboration in the Trump Era,” the lunch was a classic loyalty test, designed to see whether powerful people would demean themselves to win the president’s favor and avoid his wrath. The test works because it forces people to choose between public integrity and private comfort. If you pass the loyalty test, Weisberg says, you fail the backbone test. “The only way to resist this kind of thing is by acting collectively,” he said. “If all of those guys had just said, ‘We’ll agree to the terms of the self-regulation, but we don’t accept the name change,’ what would he have done?” The implied answer is nothing. A president cannot actually prevent a company from saying “artificial intelligence.” He cannot arrest the English language. The word will not disappear from academic papers, open-source code, or corporate websites just because an executive order declares it forbidden. The one thing the order can really take from those men is their nerve. And they proved, without saying a word, that it can.
This was never really about a name. It was about language and power. Trump has already shown how much he enjoys renaming the world around him: he has altered the names of gulfs, lakes, and centers for the arts, usually with no more justification than a whim. But artificial intelligence is not a body of water or a building. It is an entire discipline, an entire culture, an entire history. Words are how people think. A term like “artificial intelligence” is not merely a label for a set of algorithms. It contains a worldview: that intelligence can be designed, that it is artificial rather than natural, that it is something we build in our own image. “Super Intelligence,” by contrast, is a slogan. It sounds like a comic book villain’s dream or a motivational poster in a startup. It implies a power that is already beyond human comprehension, which makes it easier for the few companies that control it to justify their secrecy and harder for the public to hold them accountable. Trump may not have intended any of that. He was probably just annoyed by the phrase, as he is annoyed by many things he doesn’t fully understand. But his instinct to rename the field is not a harmless quirk. It is a way of erasing the past, and the silence of the AI elite is what makes that possible. For seventy years, “artificial intelligence” was an invitation to wonder, a question posed by scientists and dreamers. “Super Intelligence” is a decree, a command to bow before an unspecified future. If you let the most powerful man in the world replace the founding term of your profession, then every other battle over definition, safety, transparency, and responsibility will be decided on his terms too. What happens when someone tells them that “alignment” is forbidden? That “bias” is forbidden? That “openness” is forbidden? The word is always the first thing to go, and it is usually followed by the thing itself.
Could a name change kill artificial intelligence? No. The technology will continue. Dario Amodei, Jensen Huang, and company are not going to stop spending billions of dollars because a sentence on White House stationery told them to adjust their vocabulary. The chips will still be built. The models will still be trained. The chatbots will still be released, though their marketing copy might now say “SI-powered” and sound like a bad science-fiction trailer. In the long run, the name may not matter at all. What matters is what the willingness to surrender it tells us about the people in charge. This is the part that should be genuinely worrying. The AI leaders have spent years saying that artificial intelligence could be the most important technology in human history, that it could cure disease, solve climate change, eliminate poverty, or destroy everything. They have also insisted that they alone can be trusted to develop it safely. But if that confidence were real, they would have found a way to say “No, Mr. President.” They would have realized that the name of their field is the castle wall, and if they won’t defend the outermost wall, what will they do when the real siege begins? The next fight might not be about words. It might be about export controls, compute caps, liability rules, or the decision to pause giant experiments. If the leaders of the field cannot even defend their own vocabulary against a politician with a napkin and a grievance, they will be much less likely to defend the public against their own worst product. Or to defend themselves when a future government decides that it owns not just the phrase, but the technology itself.
And yet, in a strange way, the old name may outlast the decree. Language is not as easy to control as politicians think. You can command federal workers to write “Super Intelligence” in official reports, but you cannot command the English-speaking world to forget that “artificial intelligence” was there all along. There will still be books, graduate courses, conferences, and reams of academic papers with the old words stamped all over them. The term is woven into the code, embedded in the names of research groups, and carved into the syllabi of a thousand university departments. There will be a generation of scientists who continue to say AI out of habit, or stubbornness, or love. It may eventually fade from official language, but the history of science is full of words that survived emperors and politicians because the people who do the work know that naming is not the same as understanding. The real problem is not that we will lose the phrase. The real problem is that the men who could have defended it chose not to, and we all watched. It will be a long time before we know what that silence cost. It may turn out to be nothing, just a joke, a story that the executives tell at Davos about the time Trump made them call themselves SI lords. Or it may be the moment they revealed that they are not leaders at all, only managers. The machines they build are getting faster every day. The words they use to describe them are now dictated by people who will never understand them. That is not artificial intelligence. That is just intelligence, surrendered.
