Is AI the same as superintelligence?
No. "Super Intelligence" is the US administration's new name for AI in general; "superintelligence" in AI research means something that does not exist yet. Here is the difference, in plain English.
In short: No. Since September 2026, “Super Intelligence” (SI) is what the US administration calls artificial intelligence in general. In AI research, “superintelligence” has a much narrower meaning: a system that greatly outperforms the best human minds at nearly every task that matters. Nothing that exists today meets that definition. When you see the word, the first question is which meaning is intended.
The three words, in order
Think of a ladder with three rungs. The words are used loosely in the press, but each has a reasonably settled meaning among the people who work on the technology.
| Term | What it means | Does it exist? |
|---|---|---|
| AI (artificial intelligence) | Any software that does tasks we associate with intelligence: recognising speech, translating, writing, playing chess, spotting a tumour on a scan. | Yes. It has for decades, and today’s chatbots are the most visible form. |
| AGI (artificial general intelligence) | A hypothetical AI that could learn and perform roughly any intellectual task a human can, rather than being good at a narrow set. | Disputed. Some researchers say today’s systems are approaching it; others say the gap is large. |
| Superintelligence | A hypothetical AI that greatly exceeds the best humans at nearly all intellectual tasks — science, strategy, persuasion, engineering. | No. It is the subject of forecasts and safety research, not a product. |
Superintelligence, in the research sense, is “an intellect that greatly exceeds the cognitive performance of humans in virtually all domains of interest.” That is the philosopher Nick Bostrom’s definition from his 2014 book, and it is the one most researchers use.
What changed in September 2026
The US administration began using “Super Intelligence” as a replacement name for artificial intelligence as a whole. The sequence was a Truth Social poll on 19 September proposing “Superior”, “Extreme” or “Supreme” Intelligence; a second poll on 21 September between “Super” and “Superior”; and a statement at the UN General Assembly on 22 September that US government documents would use “super intelligence” from then on. The full dated timeline is here.
Under that usage, a customer-service chatbot, a spam filter and a self-driving car would all be “super intelligence”. That is a naming decision about existing technology. It does not mean the technology has become superintelligent in the research sense.
Why the distinction is worth keeping
Three reasons, none of them political.
Forecasts depend on it. When a researcher says superintelligence might arrive within a decade, or might never arrive, they mean the narrow definition. If the word is also used for what already exists, the forecast becomes meaningless.
Safety arguments depend on it. Much of the debate about AI risk is specifically about what happens if a system becomes far more capable than its makers. That is a different question from whether today’s chatbots are reliable, and it needs its own word.
Your own decisions depend on it. A business asked whether it should “adopt super intelligence” needs to know whether it is being asked about tools it can buy this year or about a hypothetical future. A parent reading that a school is “using super intelligence” needs to know it means ordinary software.
How this site uses the words
- AI for the technology in general use, including chatbots, image generators and coding assistants.
- Superintelligence only in the research sense: a hypothetical system beyond human ability.
- “Super Intelligence” (in quotation marks, capitalised) when reporting the US administration’s usage.
If we slip, tell us.
Where to go next
- What is superintelligence? — the research meaning, in more depth.
- Glossary — the other words you will meet, defined.
- The rename timeline — what was said, when.
Sources
- Nick Bostrom, Superintelligence: Paths, Dangers, Strategies (Oxford University Press, 2014), chapter 2.
- Reuters, “Trump says US will henceforth call AI ‘super intelligence’”, 22 September 2026 — via 933thedrive.com.
- TechCrunch, “Trump says it’s time to rebrand AI with a new name”, 19 September 2026.
- IBTimes UK, “Trump Wants To Rename AI ‘Super Intelligence’”, 22 September 2026.
Common questions
Is ChatGPT a superintelligence?
Not in the research sense. ChatGPT and similar systems are very capable at some tasks and poor at others, and they do not exceed the best humans across nearly all domains, which is what "superintelligence" means in AI research. Under the US administration's 2026 usage, however, all such systems would be called "super intelligence".
What is the difference between AI, AGI and superintelligence?
AI is any software that performs tasks we associate with intelligence. AGI (artificial general intelligence) is a hypothetical AI that could do roughly any intellectual task a person can. Superintelligence is a hypothetical AI that greatly exceeds the best humans at nearly all such tasks. Today's systems are AI; whether they are approaching AGI is disputed; nothing today is superintelligence.
Why is AI being called super intelligence?
Because President Trump said on 22 September 2026 that US government documents would use "super intelligence" instead of "artificial intelligence", after proposing on Truth Social that the old term was "inaccurate, and very ineloquent". No other government or standards body has adopted the term.