Media and ai education through creativity

What if artificial intelligence became a tool for developing critical thinking? Account of a training sequence that proposes to rethink media education through creativity, humility, and collective experimentation.

A reimagined training for cultural professionals

This session was part of a “Education of the Gaze: Images, Media, Information” training program, entirely redesigned by Florian Lecron (training coordinator) for Images en bibliothèques. For three days at the Bibliothèque nationale de France, about fifteen librarians from all over France, from Clamart to Lorient, from Neuilly to Épinal, gathered to deepen their skills in media and information literacy. The program was ambitious and diverse, connecting contemporary digital issues with fundamental questions of media education, to the great satisfaction of the participants.

My contribution focused specifically on artificial intelligence in the field of media education. Before diving into the technical subject matter, I wanted to establish a preamble that I consider essential. Because media education is not a trivial subject, much less a territory where one can proceed with ready-made certainties.

The participants represented a beautiful diversity of institutions and experiences: some managed audiovisual collections, others already organized media literacy workshops with schoolchildren, still others were in charge of digital projects in their media libraries. Yet all shared the same expectation: to find concrete tools to engage their audiences, especially young people, in critical reflection on images and information.

The democratic paradox of critical thinking

First and foremost, I wanted to establish the true stakes of media education. We often speak of critical thinking as something obvious, as a noble and consensual objective. Yet the reality is much more complex and demanding. Critical thinking is above all the capacity to think for oneself. And this thinking for oneself paradoxically represents a great social danger, because it can lead to exclusion from the group. As Hannah Arendt emphasized in The Crisis of Culture (1972), critical thinking implies a withdrawal from the common world in order to better return to it.

Thus, I had brought a certain number of books, placed on a table in the middle of the room, which they appropriated at the beginning of the session, which they photographed, and whose photos they shared in a very simple digital interface that I had created for them. This is also a mediation tool that I thus transmitted to them, through their own experience. They would later upload their own films to this space autonomously, before the final screening.

Let’s return to “critical thinking” - this word that we pronounce with such ease is actually extremely difficult to implement. The greatest difficulty lies in the fact of assuming one’s disagreement with the thinking of others, even if they are in the majority. This is not a disembodied intellectual exercise, but a genuine act of courage that engages our social position and our sense of belonging.

This difficulty is shared as much by the participants as by the person facilitating. Despite the very subject that is ours, we, as facilitators, can find ourselves afraid of being destabilized. And not accepting it. And thus finding ourselves taking power over people by imposing our thinking on them. Even though the subject of media education is precisely the opposite of that. This is why I consider media education to be one of the most difficult and demanding educational subjects to conduct.

Humility as a pedagogical prerequisite

I emphasized to the librarian trainees a point that may seem provocative but that seems fundamental to me: those who believe, in the field of media education, that they are right, that they have a good word to transmit, well, it is precisely they who would need to be educated in media literacy! Because they are, without realizing it, at the radical opposite of the democratic values they are supposed to transmit.

This preamble is not demagogic discourse. On the contrary, it is a discourse of democratic rigor, which one must apply to oneself first and foremost. And it is a very difficult exercise, let’s not hide it. Establishing this preamble seemed essential to me so that openness would be truly present in our approach to media education, in our ways of being together, which is the basis of the possibility of working together, and so that respect for thoughts different from mine would be at the heart of our approach.

Without this prior awareness, we risk creating a social space opposed to what we would like to defend. The history of popular education is full of examples where the best intentions transformed into new forms of symbolic domination. Pierre Bourdieu and Jean-Claude Passeron brilliantly demonstrated this in The Inheritors (1964): every educational action carries within it the risk of reproducing the inequalities it claims to combat.

TikTok, or the radical exercise of openness

To concretely illustrate this requirement for openness, I took an example that never fails to surprise: I stated before the participants that TikTok was a wonderful space. I know perfectly well that this is not at all the dominant received idea, and that’s precisely why I chose this example. It was about deliberately taking the opposite view of the received idea, so that we could share together the openness that must be our standard with respect to ourselves.

If we ourselves have not opened ourselves to what disturbs us or to what we may consider dangerous for society, how could we claim to educate others? Shouldn’t we first work on our empathy to realize that perhaps those who use TikTok think the same thing about us, that we do about them? That perhaps they too find us dangerous for society and for the world because we don’t share their beliefs, their cultural practices, their modes of expression?

Someone who thinks they are right because they are, for example, secular and that those who are not would be inferior beings mired in error, demonstrates an obscurantism that disqualifies them de facto from any action in media education and construction of critical thinking. Because they have no critical thinking about themselves.

From theory to practice: entering generative AI

After this theoretical and ethical preamble, we entered the heart of the matter: generative artificial intelligence. I had initially planned simply to facilitate the workshop, in the same way as the one I had conducted at the Forum des images with the young people from the Tumo Paris school. But I quickly realized that the librarians’ needs were broader. It was not only about having them live through a creative experience, but also about transmitting to them conceptual tools, critical reflections, and distance from this subject, its tools and its concerns, far beyond a reproducible workshop model.

The workshop I usually offer to Tumo teenagers follows a precise process in several stages: using AI to imagine a scenario, generating prompts to create images, creating visuals with tools like Playground v2.5 via the Pœ.com application, then video editing with the CapCut application. Everything takes place on the participants’ mobile phones, thus democratizing access to these technologies. The proposed themes revolve around the future, the risks and promises of AI, or future professions. The mise en abyme of questions around AI, with AI tools, seems particularly fruitful to me.

For the librarians, I adapted the formula. Rather than having them immediately move to creation, I first took the time to explain the historical and technical context of generative AIs. We discussed Ada Lovelace, pioneer of programming, the “uncanny valley” theorized by Masahiro Mori, cinematic representations of AI in Tron, Terminator or The Matrix, the notion of technological singularity, deep learning, Turing tests, Asimov’s laws of robotics, the victories of Deep Blue and AlphaGo, etc.

Creativity as a method of de-dramatization

Then, I engaged them in the creative process. We explored all the elements that constitute a film, and I emphasized the point that what matters is not so much the “mastery” of a technology, but what it can be put in service of. Because by clicking on a single button, you can have AI create a video, but on the other hand, making a film that resembles us, that transmits a point of view on the world, is a commitment to the details of a process. Therefore, the most important thing is the process, the method, the way of engaging meaning and sensitivity in what we do.

So we specified the elements of a film and the stages of its production: writing, storyboard, montage of still images, application of AI animations to create movement, and then recording voices in an embodied way, to rehumanize the film.

Then, I gave them a complete demonstration of the technical process, and they were immediately very motivated and very autonomous. In a short time, they managed to create very diverse films, each exploring different aesthetic and narrative directions. Some imagined dystopian futures, others created poetic clips, still others developed documentary narratives on the impact of AI in libraries.

What particularly struck me was the speed with which they appropriated the tools. While I expected to have to accompany each pair very closely, I found that after the initial demonstration, they experimented freely, tested different approaches, advised each other. Creative autonomy was achieved.

I had clearly explained to them that what mattered was the journey, the importance we give for ourselves to what we’re going to do. A film is composed of an enormous number of elements, and one must absolutely not remain on the technical level but really enter into the content, into the meaning of why we do things. This approach echoes what Célestin Freinet developed with his techniques of free expression: technique is only a means in service of a creative intention and personal questioning.

Critical issues revealed through practice

The strength of this approach through creativity is that it de-dramatizes the subject of media. Rather than holding a moralizing or anxiety-inducing discourse on the dangers of AI, we allow participants to experiment with it concretely, to understand its mechanisms, possibilities, and limits. And yet, this method allows for building genuine critical thinking, that is to say, thinking for oneself, anchored in lived experience rather than in prejudices or abstract fears.

During the workshop, several critical questions emerged naturally from the practice itself. Where do the images generated by AI come from? What are the issues in terms of copyright? What is the ecological footprint of these technologies? How do these tools shape our creativity? These questions were not imposed from the outside, but arose organically from the creative process itself.

One participant notably remarked that, despite her efforts to create diverse characters, AI tended to reproduce certain aesthetic stereotypes. Another noticed the difficulty in obtaining visual coherence from one image to another, which led him to reflect on the question of identity and narrative continuity. These empirical observations then nourished collective exchanges on algorithmic biases and the current limits of these technologies.

The moment of the final screening of the films was very important: everyone had gone all the way, and showed others their work, of incredible diversity, in so little time and with the same instructions and the same tools.

Transmitting rigor rather than certainties

At the end of the half-day training, I left convinced that our role, as media educators, is not to transmit certainties but rigor. The rigor of openness to others. The rigor of questioning our own prejudices. The rigor of complex thinking that refuses ready-made answers.

The librarians present left with new technical skills, certainly. But I especially hope that they have integrated a posture, an ethics of media education that places humility and dialogue at the center of practice. Because it is this posture that will make the difference in their future workshops with the audiences of their media libraries.

As Paulo Freire wrote in Pedagogy of the Oppressed (1968), “no one educates anyone, no one educates themselves alone, people educate each other, through the mediation of the world.” This training will have been, I hope, a space of co-education where everyone, myself included, will have learned from others. A space where we collectively explored how artificial intelligence tools can serve not to impose a vision of the world, but to open spaces for shared questioning and creation.