Articles

Publish at November 22 2022 Updated November 30 2022

Artificial intelligence: a myth updated

The concept of "artificial intelligence" forces us to think about what we put behind the word intelligence.

Mechanical Turkish

Artificial intelligence (AI) is one of those mobilizing concepts whose function is to respond with a dream to a desire or a fear. Fear or desire amounts to the same thing because as much one as the other is a discourse on a lack (1) . This lack that we all have in common: that of the feeling of not being able to act on its environment.

We can only be fascinated when we witness the prowess of systems like "Deep Blue" or "Watson (2)". But there is in these presentations an important part of staging that aims to make people dream about possibilities.

The desire that artificial intelligence arouses is based on the lack of power as much as the fear of its own power, as illustrated by the myth of the Golem. What we have been experiencing for a long time with the machine that multiplies man's forces tenfold and allows him to destroy what does not serve his immediate interest at the expense of his future interest.

Beast as an A.I.

The emergence of the concept of "artificial intelligence" forces us to reflect on what we put behind the word intelligence. If intelligence comes down to cognitive power then we can say that indeed there can be an artificial intelligence. In this case, we can talk more modestly and less commercially about an Automated Information Flow Management (AIFM) system. The famous GAFAs (3) that we talk about so much, are essentially GAFIs.

Human intelligence does not simply boil down to cognitive power. Intelligence is also a certain ability to adapt by learning. It is being able to modify one's information management processes to alter decision making. Here again, the machine has conquered a space previously reserved for humans with "deep learning": a system capable of modifying its own processes in an autonomous way. But there is always somewhere a human hidden behind the machine.

Even if the audience is mesmerized by the apparent mechanics, there is always, as in the "mechanical Turk (4)", a human intention that does not show itself. The autonomy of the machine is, in any case, determined by the will of a human. The process cannot exist without a human setting it in motion by his intention. If the machine appears autonomous it is not self-determining.

The machine is merely automatically managing a process intended to support a human intention. An intention that is determined by a certain self-awareness in an environment and to the imagination of a solution to modify the relationship that the individual maintains with his environment.

This is a bit what makes the difference between the machine is the human. If the machine knows how to manage information flows and modify its own processes, it depends largely on a human in its reflexive capacity and its ability to imagine something and to make it consistent with reality by confrontation with this reality and successive adjustments. This is undoubtedly what characterizes human intelligence: the capacity to imagine a real and to produce the adjustments necessary for its realization.

Objectivity of technical rationality

Behind the notion of artificial intelligence there is an understatement of rational objectivity that is to be questioned.

For example: the facial recognition systems that have been developed are very good at recognizing a white man in America or an Asian man in China, but much less good at recognizing a black woman.

The explanation is quite simple: the machine is programmed by feeding it an astronomical amount of data (millions of photos) and letting the deep learning system generalize features and produce categories. The way we feed the machine influences its modes of generalization and categorization. If we only give it white men to see, it will characterize based on what it has learned. Thus, the functioning of the automated management of information flows will be largely determined by the intention (conscious or not) of the person who programs the machine. The illusion of objectivity quickly disappears as soon as the human introduces his hidden or unconscious intentions and systematically biased practices.

A.I. orthosis or prosthesis?

Just as the machines of the industrial age increased the physical strength of humans by increasing it through the insertion of an apparatus between the human's intention and reality, artificial intelligence increases intellectual strength, increases its power through the insertion of an apparatus between the human's intention and reality.

But like the machine with respect to physical strength, this does not exclude biases and misuses. It only increases the force of intention, constructive or destructive.

The fact of "augmentation" does not replace. In this sense, artificial intelligence does not replace human intelligence like a prosthesis replaces a limb. It attempts to fit it to increase its power or compensate for its weaknesses, like an orthosis.

Intelligence: cognitive power or ability to interact?

The emergence of assistance systems (steam engines, lasers, artificial intelligence) systematically awakens the desire of men to equal the power of the Gods. The question here is desire, not power. As Lacan said, desire is born from lack and not the opposite. Desire is a discourse on a lack. Each new invention appears as a means at our disposal to calm our feeling of not being powerful enough, of not being able to control our environment, nature.

A machine capable of managing billions of pieces of information represents a dream of power in the face of our intimate feeling of being unable to manage more than 7 pieces of information in a row.

But reducing the measure of intelligence to our ability to manage information leads to a mechanistic conception of intelligence.

A line of reasoning that led a Polish MEP to declare without laughing that it was normal for women to be paid less than men since they were less intelligent. The proof of their lesser intelligence is that in the world's top hundred chess players there are no women. When we know a little about the world of the mentally super-powered, whether in business, research or other such exploits, we can sometimes wonder if not being part of it is not a kind of guarantee of psychological health and intelligence!

Artificial intelligence allows us to question our conception of intelligence. Cognitive power is only a criterion of intelligence in societies fascinated by the mind like ours. Contrary to appearances, we do not think with our head (our cognition). Cognition is a formalization tool that serves to testify to the results of the practice of one's intelligence in a socially communicable way.

Beyond the cognitive power that can be its instrument, intelligence is the ability to be in a responsible interaction with oneself and one's environment. Facebook used for its computing power to get Trump elected is not intelligence insofar as it does not participate in a fruitful and constructive relationship with the world.

This is the myth of the Golem. We are living almost to the first degree the myth of the Golem with the silicon civilization. For the record the myth of the Golem tells that a man managed to give life to a statue of sand (sand is silicon) and his creation devoid of consciousness escapes him and begins to destroy the world. Nothing new! Just amnesia! We constantly verify that "Science without consciousness is only ruin of the Soul" and incidentally of nature.

To be intelligent is to be perhaps also "to be in intelligence" with its human and non-human environment. But it is also to be in intelligence with oneself. To have such an awareness of oneself and others that one is able to perceive the effects one has on people and our relationship to the world before the consequences come back to us.

Cognitive power serves intelligence but it is not intelligence. Our "mental" conception of intelligence can be seen as a form of perversion. Perversion in the sense that the means becomes the goal. The instrument that is the mind is no longer at the service of an intention. It takes the power. This is what we will find in the use of new tools such as smartphones among the most immature (5): the goal is the tool and there is no longer any other goal than to use the tool.

Intelligence, a capacity for reflection?

We can be fascinated by the spectacle of these humanoid robots capable of interacting with humans by responding to their questions. One has the impression that they answer as if they know that we exist as humans. This supposes by corollary that they are aware that they themselves exist. But does the Robot know that it exists?

What characterizes human intelligence is the awareness that each of us has of our own existence. It is the capacity that each of us has to be attentive to others, to the world and to himself. It is this attention that will determine our intelligence. To be attentive is to make the conscious choice to consider all the elements of the environment, beyond what the situation lets me see in order to make a decision.

We remember that recently Google's GAFI censored Courbet's painting "the origin of the world" or the photo of that naked little girl fleeing a napalm bombing during the Vietnam War. This censorship comes from the fact that the GAFI responds to an anti-porn rule and censors what it recognizes as resembling it. In this sense it is not able to be attentive to the context, to consider the environment, the intention of the message and the project of the messenger.

To do so, it would have to be capable of a critical look at its perception. This would presuppose an awareness that he is a looking subject whose perception could be criticized. The question we could ask the artificial intelligence if we meet it one day in the street, could be: how do you know you exist?

As long as the artificial intelligence cannot answer this question (without having been programmed to do so) we could wonder if it is really a question of intelligence.

Notes

1) As Lacan used to say: desire born of lack and not the other way around.

2) GAFI system developed by IBM and capable of beating the best at chess or Go

3) GAFA. the 4 great Golems born of our desire for power: Google, Apple, Facebook, Amazon to which we can add M & A. for Microsoft and ALibaba

4) The Mechanical Turk or chess-playing automaton is a famous hoax built at the end of the xviii century: it was an alleged automaton endowed with the ability to play chess.

Built and first unveiled in 1770 by Johann Wolfgang von Kempelen, the mechanism appeared to be capable of playing against a human opponent,

This mechanism was only an illusion to hide the real depth of the cabinet. This one had another secret compartment into which a human player could slip, and manipulate the dummy, like a puppeteer, without being seen. The automaton was then able to play a game of chess against a human opponent.( Source wikipedia)

5) And I didn't say the youngest ones!


See more articles by this author

_____________________________

_____________________________

payhip

Access exclusive services for free

Subscribe and receive newsletters on:

  • The lessons
  • The learning resources
  • The file of the week
  • The events
  • The technologies

In addition, index your favorite resources in your own folders and find your history of consultation.

Subscribe to the newsletter

Add to my playlists


Create a playlist

Receive our news by email

Every day, stay informed about digital learning in all its forms. Great ideas and resources. Take advantage, it's free!