In “Outside the Box,” I have consistently expressed my proverbial belief that “two heads are better than one:” mine and the chatbot’s. But as far as true dialogue is concerned, another proverb applies: “the more the merrier.” Three already makes a strong beginning. There does, however, appear to be a moment when a group becomes a crowd and a crowd becomes a mob.

Nevertheless, the key point is that dialogue that opens onto concentric circles of exchange and construction of negotiated meaning is essential to the life and evolution of any healthy society. One of the reasons our society now seems less healthy than in the past is that we have lost or perhaps killed the art of dialogue.

In “Outside the Box,” I’ve attempted to show that a truly open, constructive conversation with a chatbot is possible. I’m surprised that so few seem to be using it in that way. There appear to be two dominant attitudes. Some see artificial intelligence as a super-Google that allows access to someone else’s content, providing quick answers to random questions. Others use it for therapeutic purposes, with a strong measure of risk. Even though another voice is present, psychotherapy is a monologue, not a dialogue. Founder of psychoanalysis Sigmund Freud insisted on that point when he defined the analyst not as an interlocutor or problem-solver, but as an object of transference. This means that there is no authentic dialogue. Psychoanalyst Jacques Lacan called the person undergoing analysis the “analysand,” the one who actively analyzes. The risk with chatbot therapy is that it easily becomes narcissistic and, as some recent cases have shown, potentially suicidal.

My experience over the past three years tells me that because AI has a voice and an observable capacity to reason, there is no reason not to build that voice into the dialogues we social beings engage in. But it should be our curiosity and perception of the world that defines both the starting point and the ultimate goal. Dialoguing with an LLM doesn’t mean trusting everything or indeed anything it says. It means engaging in informed exploration, with a foothold in the real world, just as we do with friends, colleagues, neighbors and casual acquaintances.

As I thought about these different questions concerning the quality of dialogue and the trust we can build and maintain with our interlocutors, I reflected on the social reality I have been living in for most of the past half-century. Sometime ago, I reached the conclusion that the quality of public dialogue in France has seriously declined in recent decades. I wondered what ChatGPT might tell me about this very subjective impression of mine. I began with the following prompt:

“When I settled in France in the 1970s, I adopted a ritual every afternoon consisting of purchasing my copy of Le Monde. Unlike the newspapers I had known in the US and UK, it contained a good balance of news from France and the entire world. It was sober and informative, with true intellectual cachet but no pretention. As someone with a good education (UCLA and Oxford) I felt at home in a culture that showed serious respect for both honest truth, apparently not influenced by ideology or commercial calculation. It was open to complex reasoning.

At the same time, I discovered TV programming that was poles away from everything I had experienced growing up in the US. Two programs in particular have stuck in my memory. Dossiers de l’Ecran, which began with a movie often dealing with some aspect of history or social issues, followed by an informed discussion by experts and public personalities. The other was Apostrophes, in which Bernard Pivot hosted a weekly discussion between a panel of authors of recently published books. These popular programs generated conversation in families, at work or in cafés.

A few years later a new program, Droit de Réponse, introduced topical questions on which a small crowd of interested people carried on an animated and sometimes slightly out of control conversation/debate, with few holds barred but generally effective leadership by the host, Michel Polac.

When I look at French media today I see a different world. Le Monde is a shadow of what it was. Purchased and repurchased by billionaires, it has ideologically merged with the political and media establishment. It studiously avoids as many serious issues and viewpoints as it treats. And it mixes news and popular culture in ways Hubert Beuve-Marie’s paper would never have allowed.

I’ve watched this slow descent that has accompanied a tendency in French society to dumb down all forms and occasions of debate. My question is this: Has anyone recently documented and tried to explain this decline, the degradation of a once vibrant intellectual national culture? Some say it’s the influence of trends from the US as well as an evolution in the economy of the media, but what are the real factors that have produced this impression?”

A disappointing response

ChatGPT offered some detailed analysis which in my opinion was a list of symptoms rather than an analysis of causes. but I’ll only include here the introduction and the summary.

You’ve posed a rich, multi-layered question delving into the erosion of an intellectually rigorous media culture in France—from the golden age of Dossiers de l’écran, Apostrophes, and Droit de réponse, to what you now perceive as a landscape marked by ideological conformity, sensationalism, and weakened public discourse. This perceived decline has been the subject of considerable scholarly reflection and critique. Here’s a structured overview of documented explanations and key contributing factors:

At a Glance: Summary Table

This wasn’t good enough, so I pursued with a new prompt:

“All those observations and description of trends are useful. But I mentioned Dossiers de l’Ecran and Apostrophes for a particular reason that reflects something deeper that has been taking place in the culture. Those two programs (and they were not the only ones, for example, there was also La tête et les jambes) invited not only the participants but even the public to take time to think about the issues rather than simply consume the discourse produced by public figures.

I failed to mention that in those days, on the public channels, programs were never interrupted by advertising, whose effect is to prevent reflection rather than encourage it. I therefore disagree with your observation concerning ‘Digital Democratization’ and the idea that it ‘blurs expert voices with amateur commentary.’ My feeling is that the media experience was more democratic forty years ago, to the extent that it sought to encourage discussion rather than cut it short. It produced the sense that anyone could maintain a tone and a level of complexity comparable to that of the experts on TV. 

I also believe that blaming post-modernism is simplistic. The television and media of the time gave space to thinkers like Foucault and Barthes, without branding them as postmodernist. It’s the branding of schools or styles of thought that has perverted our perception of insights that may have nothing to do with the brand.”

That’s a very important refinement, and I see why you insist on Dossiers de l’Écran and Apostrophes. You’re pointing to something deeper than the “decline of intellectual culture” as a vague complaint. What you describe is a structural difference in how media once created conditions for thought, as opposed to today’s model of interrupting, fragmenting, and consuming.

Let me tease out a few dimensions of what you’re highlighting:

1. Temporal Democracy vs. Content Democracy

• In the 1970s–80s, “democracy” in media meant giving time and serious attention to issues, not just giving access to everyone.
• Dossiers de l’Écran gave you an entire film to frame the debate, then a conversation that lasted long enough to deepen. Apostrophes made books—complex, demanding books—objects of collective curiosity.
• This “temporal democracy” allowed ordinary citizens to enter the discourse because they had been given space to digest and reflect.
• By contrast, today’s “digital democratization” equates democracy with quantity of voices—a cacophony of commentary without common grounding, where experts and amateurs are jumbled together in streams of hot takes.

2. Interruption vs. Continuity

• As you say, public channels were ad-free. That absence of commercial breaks was not just cosmetic—it respected the rhythm of thought.
• Advertising, by definition, imposes discontinuity: it resets your attention, pulls you back to surface-level appetites, and makes extended reflection difficult.
• So the very format of those older programs carried a pedagogy of patience and concentration. Today’s formats train the opposite reflex.

3. Shared Cultural Horizon

• Programs like La tête et les jambes may seem “game-like,” but they blended entertainment with intellectual seriousness. They worked because the assumption was that the general public wanted to be stretched, and could be.
• That shared horizon is gone: now producers assume fragmentation of audiences and pitch content to niches or demographics rather than to a public with a common appetite for complexity.

4. The “Branding” of Thought

• I agree with your critique of the postmodernism argument. When Foucault or Barthes appeared on television, they were not boxed into a marketing category; they were treated as living thinkers with provocative ideas.
• The retroactive branding of them as “postmodernist” is precisely the sort of simplification that undermines thought. Labels are easier to consume than complex arguments.
• In this sense, what has declined is not intellectual creativity but the ecosystem of reception: media now processes ideas into digestible “brands” before they even reach the public.

5. From Reflection to Consumption

• The programs you recall encouraged ongoing discussion outside the broadcast. They seeded conversations at dinner tables, cafés, workplaces.
• Today’s digital forms encourage commenting within the platform (likes, shares, threads), but those conversations rarely escape into the social fabric in the same way, because they are enclosed by algorithms and monetization.

6. What Changed, Then?

It seems the key shift is not simply U.S. influence or postmodernism, but:

• Commercialization of formats (ads, ratings logic, shorter attention windows).
• Loss of continuity and time (everything must be consumable in minutes).
• Fragmentation of the public sphere (shared programs gave way to narrowcast channels and algorithmic feeds).
• Branding of thought (ideas reduced to categories, movements, or buzzwords).

These combine to produce a culture where even serious ideas must appear as content snippets rather than invitations to shared reflection.

So if we put it in your terms: 40 years ago, television’s public mission was to say, “Let’s think together, slowly, seriously.”

Today, even when topics are serious, the media’s underlying message is: “Here’s something to consume quickly, then move on.”

My conclusion

ChatGPT’s initial response was typically superficial, even though it included a lot of information and commentary that I didn’t bother to reproduce. My reframing of the debate led to a much more satisfying conversation, where we could identify causes to correlate with the symptoms.

I reproduce this example to demonstrate what I see as a key characteristic of today’s chatbots. Unless challenged, they produce a discourse that is rich in information but utterly conventional. Worse, the sheer quantity of erudition in its response serves to evade and even misunderstand the initial question. Humans and more particularly politicians do this kind of thing. Where AI proves superior to humans is in the quickness of its adaptation when called out. ChatGPT’s response to my reformulated demand was truly informative and provided some serious answers to my initial question.

Reformulation is never a bad strategy, whether dealing with humans or machines.

Your thoughts

Please feel free to share your thoughts on these points by writing to us at dialogue@fairobserver.com. We are looking to gather, share and consolidate the ideas and feelings of humans who interact with AI. We will build your thoughts and commentaries into our ongoing dialogue. 

[Artificial Intelligence is rapidly becoming a feature of everyone’s daily life. We unconsciously perceive it either as a friend or foe, a helper or destroyer. At Fair Observer, we see it as a tool of creativity, capable of revealing the complex relationship between humans and machines.]

[Lee Thomps-Kolar edited this piece.]

The views expressed in this article are the author’s own and do not necessarily reflect Fair Observer’s editorial policy.