Another Proposal to Rename AI

In this post, I suggest using Super Computer Intelligence as another name for artificial intelligence (AI). This also responds to U.S. President Donald J. Trump's suggestion that AI should be renamed Super Intelligence.

The word “artificial” comes from the Latin word “artificialis,” which combines “ars,” meaning art, skill, or craft, with “facere,” meaning to make or do. So, the word refers to making something through skill. “Intelligence” comes from the Latin “intelligentia,” meaning understanding or good judgment. This word comes from “intelligere,” meaning to understand. Put together, “artificial intelligence” can be understood as making something that can understand through skill.

The term “AI” became known through the Dartmouth Summer Research Project in 1956, organized by McCarthy, Minsky, and others.¹ I argue that the names we choose are shaped by the society and history around us.² McCarthy openly admitted that he preferred “AI” as the name for their summer project because he wanted to avoid working with Norbert Wiener or being overshadowed by him. Wiener preferred the term “cybernetics.” At the time, cybernetics, defined as the study of control and communication in animals and machines,³ focused on copying feedback systems inspired by living things and using them in electronic control systems. McCarthy, however, wanted a name that covered logic, symbolic computing, and high-level programming. Historical accounts⁴ suggest that McCarthy wanted a completely new name that would stand apart from terms such as “automata theory” and “complex information processing.” He wanted a more original name that could attract financial support. Later, the name artificial intelligence fit well with Minsky's theory called the Society of Mind.⁵ This theory says that human intelligence does not come from one single system. Instead, it emerges from many smaller parts called “agents.” These agents cannot think, feel, or understand on their own, but they can be managed. 5 They can therefore be designed so that their combined results appear intelligent, even though the intelligence is artificial.

From a philosophical point of view, “artificial” means the opposite of “natural.” For example, an artificial flower may look beautiful from far away, but it does not have the feelings, sensitivity, and appeal of a living flower when we look at it closely or touch it. We could argue that if AI had originally been called computing intelligence, computer intelligence, machine intelligence, or intelligent machinery,6 people might not have developed such high expectations for it. Instead, its impressive technology might have been understood through simple, practical uses, such as recognizing images, recognizing speech, and learning sequences of words. We can also recognize that there are many kinds of intelligence, including social intelligence, swarm intelligence, evolutionary intelligence, decentralized intelligence, biological intelligence, and neuro-symbolic AI. Using different names helps us see each kind as one among many, without assuming that one is better than the others. Each can be called intelligent if it can solve a particular problem. All these kinds of intelligence can be scaled up or down, and each has its own limits.

For example, Ant System7 is inspired by the decentralized intelligent behavior of ant colonies. It can find a reasonably good path from a food source to the nest. When ants searching for food find a source, they leave pheromones along the path they take. Other ants then follow that path, guiding the colony toward the food. This behavior has been turned into an algorithm and used in computer networks to send data through Internet nodes with less traffic or load.8 Ants can therefore solve an optimization problem using decentralized intelligence. There is no central server or “boss ant” directing the others. Each ant follows simple rules, but the behavior of the whole colony is considered intelligent. The ants work very efficiently, or intelligently, on this specific task.

Biological intelligence offers a huge and seemingly endless field of knowledge and research. For example, synaptic plasticity helps us form concepts and connect ideas by strengthening links between neurons in the brain. Dopamine levels also rise when we expect to reach a goal. These dopamine signals help strengthen useful connections so that we can reach that goal.

Another example is evolutionary intelligence. Genetic algorithms copy the way species pass on genetic information through chromosome crossover and mutation. They search for the best fit, or best solution, across many generations of reproduction. Humans may find these problems difficult to solve, but using computer intelligence to solve them offers promise. Since computers can represent and model all these kinds of intelligence, and human intelligence will always remain a major goal to understand and reproduce, why not simply call these computer-based efforts computer intelligence? This would avoid the suggestion of artificiality carried by the name artificial intelligence.

Finally, while I was writing this post, U.S. President Donald J. Trump spoke before the United Nations General Assembly on September 22, 2026, and asked that references to AI in all U.S. documents be changed to “Super Intelligence” (SI).9 However, this name does not say what kind of intelligence it means, because any kind of intelligence, including human intelligence, can be superb. Also, if SI means intelligence that goes beyond human ability, that may not be possible in every area. There will always be things humans can do well that intelligent machines cannot. For example, if we keep the data gathered on Mars or at the Hadron Collider away from machines, they will never be able to surpass humans in understanding Martian matter or bosons, respectively.

Also, today's most advanced AI algorithms need enormous computing power. So, a suitable name could be Super Computer Intelligence. This name keeps the idea that AI needs supercomputing power to work properly and perhaps perform beyond our expectations. At the same time, it preserves computer intelligence as the basic term that, in my view, should be used for this technology.

References

1. McCarthy, J., Minsky, M. L., Rochester, N., and Shannon, C. E. A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence, August 31, 1955. AI Magazine 27, 4 (2006), 12.

2. Derrida, J. and Spivak, G. C. Of Grammatology, Vol. 2, pp. 162-254. Johns Hopkins University Press, Baltimore (1976).

3. Wiener, N. and von Neumann, J. Cybernetics or Control and Communication in the Animal and the Machine. Physics Today 2, 5 (1949), 33-34.

4. Radanliev, P. Artificial intelligence: reflecting on the past and looking towards the next paradigm shift. Journal of Experimental & Theoretical Artificial Intelligence 37, 7, 1045–1062.

5. Minsky, M., & Lee, J. The Society of Mind, Vol. 1. Simon and Schuster, New York (1988).

6. Turing, A. Intelligent Machinery (1948).

7. Dorigo, M., Colorni, A., and Maniezzo, V. Distributed optimization by ant colonies. In Proceedings of the first European conference on artificial life (1991), pp. 134-142.

8. Di Caro, G. and Dorigo, M. AntNet: Distributed stigmergetic control for communications networks. Journal of Artificial Intelligence Research 9 (1998), 317-365.

9. UN News. Trump defends military action against Iran, Venezuela, and calls for rebrand of AI. United Nations.

Mario Antoine Aoun is an ACM Senior Member who has reviewed articles for ACM Computing Reviews since 2006. He has more than 25 years of experience in computer programming and holds a Ph.D. in Cognitive Informatics from Université du Québec à Montréal in Canada. His main research interest is modeling memory using chaos theory and spiking neurons.