L'IA nous rend-elle bêtes ? | Les idées larges | ARTE

Chadjepeté has been embarking on the 30th November 2022 and since then nothing has

been changed.

Perhaps in a few centuries, the book of history will compare this date to the invention

of the Primerien in 1450, because the degenerative acts that create texts, images or videos

are the main focus of our society, the way we work but also of our reflection.

Each month, the book is still on the subject, but one philosopher particularly

holds my attention.

He calls Analon Bère and he is the master of conferences at the University Pari 8.

In his book of Artificial Beauty, he analyses the consequences of Lya on our thought.

For him, the term Tromper of Artificial Intelligence, we are passionate about the effects

of the disaster of these tools on our mental capacity.

He offers us the dearly referred to by a parallel to the Virgin Mary.

The invention of the writing of the ancient era, which also constituted a more cognitive

rupture in the history of humanity.

So why do we need to be aware of this artificial intelligence term?

What are the impacts cognitive of this technology on our humanity?

Lya gives us such a bet.

It is with the conventional agents, such as the Chadjepet, Mistra, or Claude,

that Lya is now concrete for the great public.

But his story lies in the middle of the 21st century in the United States.

We generally retain two founding events to date Lya's birth.

First, an article from Alan Turing, you know the name of this mathematician,

which has echoed Enigma, the Nazi-Crypted machine.

And he also created a text pioneering in 1950, which proposed to contribute to the computer

the thought of the fact.

Then it is the time of the conference of Dartmous in 1956, which made scientific

American interesting in the form of a system that the term of artificial intelligence

was really invented.

The starting point of the Annalombaer is precisely to criticize the employment of this term

as an ideological and not only scientific.

I really write this term artificial intelligence in the discourses of ideological ideas.

In reality, it is a term that has a vocation, especially, which has had historical vocation

to take financing to build a machine-type hotel, which was supposed to reproduce

the human reasoning, which was supposed to reproduce the language, which was supposed to reproduce

the game, the fact of speaking of artificial intelligence, that often invites us to compare

the performances of the machine to the performance of humans, to put in competition

actually the algorithmic machines and humans.

Their history, the world's champion of failure, has been beaten up in very singular combat

by a machine.

He was certain for time to come to the end.

During the intelligence to let his place to the formidable power of calculation,

I touched in the B&M laboratories and implanted in the entrance to an ordinate of a ton and a half.

What can induce humans to think that they have passed performance, that they will be replaced,

etc., so that they have not been too demanding on their work condition, their salary and all that.

So you see a certain way of speaking can actually have political games.

At some point, where a union can write a mail as a human, the comparison is not only

rhetoric, it is also real, no?

Yes, but that's the principle of automation.

What makes me problematic is that, in reality, human itself has always been

already supported by technical and technological and industrial protection to think in a certain way.

We do not think of his support, his trace, his image, his writing, etc., etc.

So the artificial intelligence in this sense, it does not exist, because it is a computational automatic,

or it has always existed since we are, to be, secretly, prosthetic, and then, here,

to the extent that we have to go to the chat.

The question that arises for me is not to compare humans to the machine and to know who is the best,

but it is to understand this relationship between humans, humans, individuals and groups,

and their technical and their technical tests, and to see if they are prosthetic, they are the proletarians,

that is, they are deposed of their knowledge, and also against their support, their capacity,

if they are collective.

As you have mentioned, the artificial intelligence is a promotional term,

so what would be the term, in particular, the most adapted generic to all these technologies?

So I proposed a term that is the term self-numeric or self-computational,

because it allows us to re-inscribe, in fact, these technologies, in another genealogy,

in fact, which is the genealogy of self-automatization.

So, for example, at the time of the first and second industrial revolution,

we automatized a certain number of savoir-fers in the machines that, in fact,

made, say, at the place of open-air, a certain number of gestures, etc.

And what happens today in my life is an automation of savoir-pens,

that is, instead of deciding whether to go to see such or such content from your singular memories,

you are going to confirm to the indication that you give the algorithm a recommendation.

Even if you are going to explain it from your memory and your singular imagination,

you will confirm to the indication that you propose, for example, the GPT.

Speaking of savoir-pens, Ann Alonbert can be an expression that understands intelligence.

That is, therefore, also a practice, a training, at the same time as savoir-fers.

I think that the fact of considering this technology as a digital or computational automaton,

will invite us to pose another type of question.

For example, can we automatize all savoir-pens?

What are the advantages of automatization of a certain number of cognitive, psychical, psychical, etc.

And these questions, we can't pose them if we continue to talk about artificial intelligence.

And if we focus on this tendency to anthropomorphize technological and industrial devices,

it takes for double-defeated humans in a certain way.

And in fact, we entertain with this technology, a kind of fantastic relationship,

in reality, when we talk about artificial intelligence, commercial agents, automatic mathematics, etc.

anthropomorphism, that is, the fact of assimilating human machines with smart terms or neurons

to not be seen more clearly.

On the contrary, it pushes us into a toxic comparison with machines.

We often stay in this imaginary robot that describes and critic the philosophy of the world,

which is a technique that is written in the 50s and 60s.

That is, this representation of the machine as double-defeated humans,

which will be able to become the enemy of the human world or the human world.

Open the pod bay doors, hell.

What we do, if I understand well, is that, in fact, these machines are very powerful,

these technologies are very complex, they are very common, like black wood,

which are actually quite fascinating, at the same time, and very fascinating.

Exactly. The problem for Simon Don is that,

consciously, technological or industrial functions are masked in a black wood,

and that the user or the consumer has access to the result,

and well, there is a problem of intelligibility of the device.

And so, according to Simon Don, we are in a position of consumer,

in a certain way, which is common, but which do not understand how the devices work.

And for him, that's a source of alienation.

That's why he is also on the importance of the technical culture and the understanding of the devices.

To understand a technical object and to have a right and right eye in the eye,

you have to have to know how it is constituted in its essence

and to have assisted to its young or directly when it is possible,

or by the teaching.

So, there is no teaching of the technical history, it is extremely regrettable.

You would show that there are many philosophers who think of this theory

between the morphic of the machine.

George Cangilian, who was a philosopher and a medicine scientist.

Yes, so I quote a text from Cangilian in 1980,

it's a conference called Servo and the thought, which is extremely interesting,

and in which he identifies a certain number of mathematical metaphors,

for example artificial intelligence, artificial brain, etc.,

a thought machine, saying that these metaphors, just anthropomorphic,

are to hide the presence of the decider behind the machine's anonymous image.

So what he means by that is that when we talk about an intelligence of thinking,

of algorithmic consciousness, in reality, we mask the fact that there are humans who train

a certain number of algorithmic systems, according to a number of values,

according to a certain number of economic and political objectives.

And so, again, we depolitize the technological question.

You see, anthropomorphization is a strategy of depolitization

in reality, of these technological dispositions.

The metaphors comparing the eyes to human faculties depolities are dangerous,

by doing as if these tools were our natural extension.

But it's not just marketing.

It's the same conversion agent design that is built for us to reach millions of people.

It's something that must be noted, the fact that most of the chatbots

are the first person to sing, say, young.

So, obviously, it incites the users to trust themselves, to fall in love, etc.,

to ask them all kinds of questions about all kinds of domains.

The fact that the chatbots often invite us to ask us a new question.

So, we launch a permanence, we confirm ourselves in our ideas,

saying that our questions are really formidable, original, incredible, etc.

And it's potentially very dangerous because it can create emotional dependence.

There is, beyond theory, perhaps a certain relation to the reality of the technique

that is a partially effective and emotive relationship

and that does not have to be the one that is going to have a ridiculous relationship,

that we must be neither too passionate for tactical objects,

nor too passionate for a single one, of course,

or not completely indifferent to true things and considering them as slaves.

We need a half-day life, a society with which

the current situation is correct

and perhaps something a bit astable.

Even if there is a competition for intelligence,

we see that it does not play a serious role with this technology.

This link is also effective,

and it's precisely what makes this more dangerous tool.

You may have seen a study that was published by researchers at MIT

that showed that when some of the mobile devices at the GPT,

there were certain areas that are associated with the

mental integration and the creative idea that was not a solution.

The study by MIT, which is published in 2025,

participants were distributed in three groups,

some used a type of GPT,

others with a simple search engine like Google,

and the last group was not available without a tool.

The researchers then observed the activity of the cerebral,

and the results were consistent.

The group's main goal of using LIA shows a very marked basis

of the activity of the brain,

that are in terms of attention, coordination or conceptual capacity.

The conclusion of the study is that there is a cognitive debt

for people using LIA.

Basically, the tools like Cloud or GPT allow us to do so.

Visit the active at the same time of our brain.

A bit like a sprinter that will see a robot train at its place.

I think that is quiet because when we explain,

we mobilize our memory and our imagination.

And this memory and this imagination

are basically singular.

Because all living have a different life.

In every cultural field,

whether philosophy, science, art, cuisine, sports, etc.,

the culture evolved from the abysmal vacation,

from the improbability, from singularity.

And these are the singular abysmal vacation

that allows the culture evolution.

The

So if we eliminate all the abysmal vacation,

all that is out of the form,

and if individuals and groups

are more focused on their memory and their singular imagination,

but the laws have a probability of massive amounts of data,

we have a risk of standardization and standardization

of a collective culture.

There is another entry

of this impact on our collective individual intelligence.

It is the question of the prompt, so you can talk.

So the prompt is the command that we address

especially by a LGBT chat.

What is the cognitive effect of the prompt?

The problem is that, in a way,

we still have an external response to each time,

we have a consumer position, we pass a command

and we expect a result that we arrive almost immediately.

The risk, indeed, is to reduce

all artistic activity

to a common activity of a music.

And so, for example,

we will be able to produce music,

we will be able to produce images thanks to these systems.

But the problem is that, often,

when an artist produces an image,

when he produces a music,

he does not pass by language.

He does not have the idea of the music

that he wants to produce or the image that he wants to produce.

The idea is not there at the start.

The idea is the abutism of the process.

What is the operation of a metamorphosis of the creation?

Exactly, we transform into a surprise,

in reality, by this product.

So there is the idea that everything could be passed

by a sort of linguistic command,

formed by a man,

which seems problematic to me.

Listen, I am told that we are in a very serious situation

where all human creation can be made by a union.

For example, every day,

75 thousand products in Paris are put online on the platform of 10 hours.

And the German writer, Anadamazio,

would like to have a link to write his next novel.

I started using it for the creation.

I am so happy to see the ability to create

on imaginary universe,

on things that really relate to a work of science fiction and fantasy.

In fact, the problem, if you want,

is that these are systems that make probabilities.

But truth is nothing to have with any probability.

The truth is once again the novelty.

And then the collective certification by fathers

that will judge whether or not

they know how to stabilize,

in such a theory, at such a moment.

That's why truth is not something universal.

It's something that evolved with time,

in the scientific culture.

But now, it's a probabilistic system.

So they make the opposite of the truth.

If you want, they give you the majority opinion,

stereotypes, and truth is precisely what comes to the end of the story.

The worst thing is that Lyia also ends up by sabotaging herself

because the exponential production of content generated

creates a number that is not online.

We already have the massive generation of quality content

that will be in the internet, the social networks,

the famous AI-slope,

and that will make it disappear in a mass of incine content,

potentially, relevant or important content.

In clear, of the noise, of the dogs,

a huge, vomited food that takes the form of a fabric image

in mass to hide from clicks, views, and of course, money.

And this strategy, I wrote it in the book

as a sort of industrial production of incine.

And when I talk about betaism,

I really talk about this issue of incine.

That is to say that more we use this device,

more we produce automatic content

that will then serve them to produce automatic content.

That's what we call today the model autofagy.

A little bit like if you make a photocopy of a photocopy

of a indefinite way,

at the end you will have a degraded image

because you never had a renewed in a certain way

the original if I can say.

The areas are so deep in our society that we have the impression

of being incapable of reflecting with discernment.

What I have adored in the book of Anna Longberg

is that it proposes to do it with a totally unexpected parallel.

In the remodeling of writing in the ancient Greek language.

What interests me a lot is that, in a certain way,

the question of technical and technological transformations

is that in a certain way,

it is not a question so new.

That's why I am interested in this text of Plato,

that I read in the book, which is a dialogue that is called the Fédre,

which has been highly commented by Jacques Derrida in an article

that is entitled the Plato Pharmacy written in the 60s.

And in this text, Plato is interested in writing on the writing technique.

In the book, the Plato is written on the writing technique.

In the book, the Plato is written on the writing technique.

In the 60s, the Plato is written on the writing technique.

In a context where the alphabetical writing,

which is several centuries ago,

is disseminated massively in the Greek society.

As there is, the writing constitutes a cognitive revolution.

Because Laura is a parol who is angry and who obliges to memorize everything.

While the writing allows to fix the information and transmits it in time and space.

But as all technical evolution, the writing also has negative effects.

That is, the reason why the Plato is angry is because of the fact that the writing is invented by the god Tutte,

which will have the right to teach him to say that thanks to his invention,

we can accumulate knowledge because they will not be lost anymore.

Or, this law will be the author that is not a remade but a poison.

Because when citizens will extort their knowledge in their inner supports,

they will not be able to remember themselves and therefore to exert their memory.

What is presented as a remade for the memory and knowledge

for the memory and knowledge is also presented in the same time as a poison for memory and knowledge.

And so if you want Plato in the text,

this technique of writing, under this concept of pharmacism,

which is in Greek and is a term which means that the poison and the remade are the terms that gave the term of pharmacy in French.

But in his article, which is called the Platon Pharmacy,

the Rida, also on the fact that this term in reality is Greek,

we often translated it by medication, but it means that at the same time the poison and the remade

and therefore, Stigler, in releasing the Rida, will finally support all the techniques

and can be described as a pharmacist, that is to say, as a violent ambivalent,

at the same time, poison and remade, when it is in particular of its effects on the spirits

and the individual spirits and collective cultures.

The practice and educational in the ancient Greece will apply the learning of writing,

the reading, the philosophy of the poysidium, the mathematics, etc.

So there is all kinds of new knowledge that will be seen today,

also from an collective appropriation of this technique and this practice of writing.

And then there are also political organisations like the police, the Greek city,

which will see the day from this new symbolic technology,

especially the possibility of publishing the laws and from the moment when the citizens

have learned to read and write the possibility for the history of reading the laws,

of criticizing the laws, etc.

But what I mean is that there was an adoption of political and social

and political technology.

In Greece, the writing will take advantage of everything, because it was taught in the citizens.

And that, for example, allowed to advance democracy thanks to the writing of the law.

It was such a collective reflection called an alarmber,

so that the read is actually beneficial.

I am actually replying to this pharmacological perspective

to think about the game of digital writing systems contemporary,

in addition to that currently the way that the collective and civil institutions

are being artificially generated,

and being basically toxic in the sense that they are technology

which will just be pushed out of a certain amount of mental psychicognitive capacity.

And so we must absolutely reflect on other digital systems

that allow us to support these mental and mental psychicognitive capacity

and also our collective intelligence, in reality.

What will make me today is that we have a certain poverty

in technological images,

and that we are expecting to believe that artificial intelligence, for example,

is taken to the Chathbot, such as the Chaget GPD, Grohok,

or other products produced by large companies that are foreign or Chinese.

I think that precisely the future of artificial intelligence

is already becoming more and more diverse in the model of the DIA.

And I want to say their fragility and their small size

and their speciality,

it seems to me that the artificial intelligence is more desirable

than the super artificial intelligence.

Democracy is supposed to be made in a collective way

by citizens and not by some private companies.

Get out of the big DIA, invent the Silicon Valley,

based on another way of imagining these uses.

An Alonbert offers to promote what is called the diversity technology.

The diversity technology is the fact of supporting and promoting

the alternative digital devices.

And so that's the reason why I have really opened up

to the idea of the algorithmic pluralism that we defended

when I was a member of the National Council of the United States

and that there was reprieve in the report of the digital data of information.

And also in the report of the Commission of Enquête on TikTok,

the idea of the algorithmic pluralism, in fact,

is the idea according to which the large platforms,

the large commercial networks should accept

that there are several algorithms of recommendation

that can be implemented on their platform,

precisely, so that individuals can actually

be recommended by entities that have the same choice

or that individuals can even parameterize their recommendations.

I often take an example, an association called Turn the Sun,

which develops a algorithm of collaborative recommendations

that are based on human judgments.

One thing that I advise is that I'm not going to accept all of them.

You can go to the extension and if you use extension,

then if you are on YouTube, for example, I'm going to go to YouTube.

There, I'm directly in the package of YouTube,

the automation of all of this.

Very cool.

The fact of developing a diversity technology

could allow us to invent models

that do not work according to the same bases.

In fact, that does not work according to the bases of the data

and the attention of effects that do not work according to consumerist

and extrativist, and that opens to new practices.

In fact, it is possible, obviously,

counterintuitive practices, democratic practices.

I think there is already thought of this term

of artificial intelligence,

which seeks to compare us to machines and therefore to depolitize the debates,

because Asia is the world's largest in our society.

They run a risk to our humanity,

by proposing to reflect and express ourselves and create our place.

An alomber invites us to review questions,

not to let them fight,

but to refuse that these uses are appropriate against collective interest.

They show that other models exist

so that artificial intelligence will really give us more intelligence.

To go further,

but sources are in description.

And to keep the ideas wide, there are other episodes.

See you soon.