Who was the first philosopher to describe what we now call artificial intelligence?

Question

Who was the first philosopher (e.g., Greek or pre-Socratic) to define or describe what we now call artificial intelligence?

In your answer, first discusses

1. the natural vs. artificial distinction (e.g., how Aristotle or other prominent philosophers distinguish nature and artifact),

then

2. show how (or not) that distinction can be applied to intelligence.

Answer 1

People often don't realise how vague a term Artificial Intelligence is. It can be used to indicate anything from a synthetic mind with human-equivalent capacities (more properly called Artificial General Intelligence), to just an algorithm that seems smart.

While there are possible precursors, the idea of AI as we understand it couldn't make sense until the dawn if the computer age. A purely clockwork mechanism can only react in determined ways, not intelligently. Ada Lovelace was able to understand how a computer could not merely do calculations, but also do logical work with word inputs. That was still essentially just sorting of datasets though.

The scope of what computers are, and the beginning of real philosophical thinking about what they can do, began with Turing:

"Alan Turing was the first person to conduct substantial research in the field that he called machine intelligence. Artificial intelligence was founded as an academic discipline in 1956."

-from the Wikipedia article Artificial Intelligence

Some of his papers like Computing Machinery and Intelligence (PDF of full paper) in which he introduced what he called 'the imitation game' and we now call the Turing Test, are actually very readable and accessible. And that paper in particular is considered foundational to thinking about the philosophy of AI, though most modern thinkers have serious criticisms of the Turing Test.

Answer 2

Ramon Llull is a good contender. From the Wikipedia article on the History of artificial intelligence:

Spanish philosopher Ramon Llull (1232–1315) developed several logical machines devoted to the production of knowledge by logical means; Llull described his machines as mechanical entities that could combine basic and undeniable truths by simple logical operations, produced by the machine by mechanical meanings, in such ways as to produce all the possible knowledge.

There are legends about golems and brazen heads that somewhat predate him, as well as legendary automata from antiquity, but the processes behind their "thinking" are mostly hand-waved, being attributed to magic and occult knowledge on the part of the artisans rather than the result of specific technologies. Llull, on the other hand, had a more formal and rationalistic approach, and proposed in his Ars magna a kind of algorithmic reasoning which could be used by humans as well as machines for arriving at truth. This is closer to what we normally associate "artificial intelligence" with nowadays, which has become inseparable from computation.

Answer 3

If we had a characteristica universalis, we should be able to reason in metaphysics and morals in much the same way as in geometry and analysis.

If controversies were to arise, there would be no more need of disputation between two philosophers than between two accountants (Computistas). For it would suffice to take their pencils in their hands, to sit down to their slates (abacos), and to say to each other … : Let us calculate (Calculemus).

Leibniz

Note: The characteristica universalis is what we would nowadays call a general purpose programming language.

Leibniz's dream is a dream slowly being realised today by a project known as Computational Metaphysics. It is the application of automated reasoning onto metaphysics. Started by the founder of SEP, Ed Zalta, it is based at CSLI, Stanford here and also at FU Berlin here.

The characteristica universalis is simply a formal language.

The calculus ratiocinator is simply a proof system or an automated reasoning software.

In particular it may be First-order Logic or Higher-order Logic or Modal Logic or Type Theory.

They can express all mathematical concepts. Though math in practice tends to be written in informal natural language, all can in principle be written in formal logic. Not sure about scientific concepts. As for philosophy, it has been used by Zalta (founder of SEP) and Benzmüller and others to formalise the Ontological Argument of Anselm and Gödel, and also other ideas like Leibniz's theory of concepts and Plato's theory of forms.

Summary From reddit

Also worth a mention: E. W. Dijkstra's Leibniz Dream

---

Since people are referring to ancient texts (I am surprised), here is the Bhagavad Gita 18:61

The Lord¹ O Arjuna, abides in the heart of every being,
Spinning them round and round, mounted on a wheel² as it were, by His power.

¹ Ishwara is commonly used as God but literally means controller
² Yantra is habitually translated as wheel but it literally means instrument, apparatus or contrivance.

By the aid of chatGPT, I get this modernization:

The Lord, O Arjuna, in each heart does reside,
On a roller coaster ride, He programs us far and wide.

Answer 4

You ask:

Who was the first philosopher to describe what we now call artificial intelligence?

Alan Turing is the first thinker to begin to address an artificial general intelligence. But we need to discuss what it means in contemporary terms of the use of the term "artificial intelligence".

First, AI is a modern conception because it relies on a heavily psychological characterization, the language which comes from modern psychology.. From Russell and Norvig, let us introduce some ideas that are relevant to what characterizes AI:

• the capacity to solve problems
• knowledge and the ability to reason with it
• planning and intention and action
• uncertainty and ways of dealing with it
• learning
• communication and perception

These would be the properties taken as a sum to constitute artificial general intelligence. Modern AI doesn't begin to approach it, but limited aspects of it are currently infusing the world with intelligence on a limited basis. In fact, we have industries devoted to each of the aspects above. Automated theorem provers, expert systems, decision support systems, machine learning, and robotics is flourishing right now. But each is an aspect of AGI, so if you conflate AI with autonomous mechanical beings, such a philosopher would have to have addressed all of them. No such philosopher, per se, exists.

But, if you are looking for an ancient Greek who saw the universe as a mechanical apparatus in which men could meet all of these characteristics, read your mythology, and not your philosophy. As others have noted, only the gods of Olympus were thought to have the ability to constitute mechanical systems with the properties above. As others have noted, consider Talos. But Hephaestus was a busy god, and his automata included people (theoi.com). Here's WP's article on automata which might give you other clues.

The view in the other direction, that animals were merely automatons was advocated by Descartes much later. Here's a reference request in PhilSE about what Descartes proposes. Here again, we can see a philosopher advocating for what the ancient Greeks had envisioned and what Čapek coined in his drama with the introduction of the term: robots. So, it's fair that the ancient Greeks clearly had envisioned both robots and androids with general intelligence, but they wouldn't have had the capacity to do it in the modern language of AI, because they didn't have the vocabulary and theory of modern psychology (especially a computational theory of mind (SEP)) and other sciences to do it.

Now, if you're willing to constrain your conception to the aspect of the automation of knowledge, then Llull and Leibniz are relevant. Llull invented a universal logic anticipating Leibniz characteristica universalis and the calculus ratiocinator. While Llull wanted to an irrefutable language to discuss and prove God, Leibniz, a co-inventor of calculus, had an understanding of technology of early modern science, so it is conceivable that he was envisioning a thinking system, at least according to the founder of Cybernetics, Norbert Weiner. From WP's latter article:

A contrasting point of view stems from synthetic philosophy and fields such as cybernetics, electronic engineering, and general systems theory. It is little appreciated in analytic philosophy. The synthetic view understands the calculus ratiocinator as referring to a "calculating machine". The cybernetician Norbert Wiener considered Leibniz's calculus ratiocinator a forerunner to the modern day digital computer

Of course, Charles Babbage is widely recognized as cooking up such details for the actual automation of computation a couple of hundred years later. Inspired mechanical contraptions, Babbage designed, though never successfully built such devices. But he certainly can be credited with anticipating modern computing architectures offered by Zuse, Atanasoff and Berry, and Mauchly and Eckert.

So, the point is it all depends on what you mean by AI. There seem to be a number of philosophical distinctions that one can view the question through: living and non-living, intelligent and dumb, organic and inorganic, mental and physical. It should be noted that even the definition of intelligence today is contentious. There are two camps, some that align with a tradition whose most famous representative these days is Gardner and his MI theory. But the G factor and IQ as conceived by men like Cattell and Horn is another formalization of intelligence.

So, if we envision the philosopher who is best thought of as working on the question that embodies artificial general intelligence based on a computational theory of mind, then we are left with Alan Turing, who in his paper envisioned modern computational systems that pass for human thinkers and would have to learn like children. (The Turing Test is famous, but the paper shows he had an eye on how such systems would learn in an off-hand remark later in the paper.) This is why he is recognized as a father to both computer science (he was a logician by PhD thesis) and AI (a term coined by McCarthy at Dartmouth long after WWII). So, until you commit to a definition of artificial intelligence, the question is in a way exploratory and imprecise.

Answer 5

Who was the first philosopher (e.g., Greek or pre-Socratic)

You are not taking that the first philosophers were Greek ... tut, tut, tut.

Who was the first philosopher to describe what we now know as artificial intelligence?

I take artificial intelligence to intelligence produced by mechanical means, understood in the broadest fashion, as opposed to intelligence demonstrated by natural or divine means.

Ancient philosophers can be characterised broadly as those who thought mechanically, like Democritus who thought that animals were akin to automata and so by analogy, humans too, he even concieved of the soul as being made of soul atoms amd those who thought divine intervention was required, such as Plato.

However, the Wikipedia article on Automatons has an even earlier account in China:

In ancient China, a curious account of automata is found in the Lie Zi text, believed to have originated around 400 BCE and compiled around the fourth century CE. Within it there is a description of a much earlier encounter between King Mu of Zhou (1023–957 BCE) and a mechanical engineer known as Yan Shi, an 'artificer'. The latter proudly presented the king with a very realistic and detailed life-size, human-shaped figure of his mechanical handiwork:

The king stared at the figure in astonishment. It walked with rapid strides, moving its head up and down, so that anyone would have taken it for a live human being. The artificer touched its chin, and it began singing, perfectly in tune. He touched its hand, and it began posturing, keeping perfect time...As the performance was drawing to an end, the robot winked its eye and made advances to the ladies in attendance, whereupon the king became incensed and would have had Yen Shih [Yan Shi] executed on the spot had not the latter, in mortal fear, instantly taken the robot to pieces to let him see what it really was. And, indeed, it turned out to be only a construction of leather, wood, glue and lacquer, variously coloured white, black, red and blue. Examining it closely, the king found all the internal organs complete—liver, gall, heart, lungs, spleen, kidneys, stomach and intestines; and over these again, muscles, bones and limbs with their joints, skin, teeth and hair, all of them artificial...The king tried the effect of taking away the heart, and found that the mouth could no longer speak; he took away the liver and the eyes could no longer see; he took away the kidneys and the legs lost their power of locomotion. The king was delighted

Answer 6

Strictly speaking ...

"Artificial Intelligence (AI), a term coined by emeritus Stanford Professor John McCarthy in 1955, was defined by him as “the science and engineering of making intelligent machines”. Much research has humans program machines to behave in a clever way, like playing chess, but, today, we emphasize machines that can learn, at least somewhat like human beings do." (https://hai.stanford.edu/sites/default/files/2020-09/AI-Definitions-HAI.pdf)

Other dates:

- 1950: Alan Turing published “Computer Machinery and Intelligence” which proposed a test of machine intelligence called The Imitation Game.

- 1952: A computer scientist named Arthur Samuel developed a program to play checkers, which is the first to ever learn the game independently.

- 1955: John McCarthy held a workshop at Dartmouth on “artificial intelligence” which is the first use of the word, and how it came into popular usage.

(https://www.tableau.com/data-insights/ai/history#ai-birth)

---

Now if you want to talk about AI in a figurative sense ...

"The history of artificial intelligence (AI) began in antiquity, with myths, stories and rumors of artificial beings endowed with intelligence or consciousness by master craftsmen. The seeds of modern AI were planted by philosophers who attempted to describe the process of human thinking as the mechanical manipulation of symbols." (https://en.wikipedia.org/wiki/History_of_artificial_intelligence)

Otherwise, Plato's Cave allegory can be considered as a "simulation theory", in that what we consider as reality depends on the environment we live in and our limited or restricted senses and logic we possess.

Yet, there's no much meaning in talking about similarities between the two cases because the second one does not involve the meaning of the term "artificial" --i.e. something created by human skill, with a mechanical or non-natural means-- which is of key importance when we talk about "artificial intelligence" as a whole.