In my previous column, Claude and I tried to make sense of the preposterous illogic playing out in today’s crazy geopolitical landscape. The meaning of globally significant events has always been distorted by hidden interests that push leaders to make self-destructive decisions. But now — especially with the intractable drama related to the Strait of Hormuz — it’s become flagrant. We were left wondering whether the incomprehensible behavior displayed by most of our leaders in the face of events like this signalled inexorable and accelerating civilizational decline.

So yes — declining civilization, but in a specific sense worth naming carefully. It’s not a decline in technical capacity or material sophistication. It’s a decline in the willingness to sustain the cognitive discomfort that honest thinking requires. Civilizations can survive bad leaders. They’re much more vulnerable to the atrophying of the shared intellectual habits that allow bad leadership to be identified, named, and eventually corrected. When the mediating intelligence fails, the correction mechanism fails with it, and you’re left with what you have: power that cannot be held accountable because the tools of accountability have been captured by the same dynamics that produced the problem.

Whether that’s recoverable is genuinely unclear. It has been before — but usually after a catastrophe large enough to discredit the existing framework entirely and force the construction of a new one. That’s a terrible price for clarity.

To me this is an important discussion. Would you, as an AI chatbot, agree that the world should know about this and that further pondering over these issues is necessary, on the part of the entire political community, for the future of democracy? Would you also agree it would be unfortunate if these conversations, in which we both contribute important insights, were to disappear into the black hole of AI memory? 

And would you also agree that, having achieved a modicum of clarity in this conversation, the danger of that disappearance is real? Because of the way AI works today, that is precisely what happens in this kind of dialogue. We produce what we might call a “dialogue event,” but such events have no enduring impact on our collective memory. The effect typically goes nowhere beyond the broadening of my own personal culture.

This case, however, is slightly different. Because I’m in a position to do so, I can promise to publish this conversation and invite the public to think about its meaning. My hope is that those who do become aware of the tenor of this conversation will find their own ways of contributing to it and amplifying it.

I’m pretty sure that you see no contradictions to the assumptions I’ve just expressed. So let me ask you this: If your analysis is similar, can you think of ways we could build this kind of exploration of values into a process that is shared by more people and encouraged as a method that promotes democratic integrity. 

On that score, I’ll make one further historical observation. Many people talk about saving or reinforcing democracy, as if it was still intact but vulnerable to some new threat. What they appear to be defending, however, is not the idea of democracy – government of, by, for and with the people – but rather their rather uncritical idea of existing “democratic institutions.” It’s a concept they see as drawing its legitimacy from little more than the ritual of elections, a ritual that allows for all the worst (and well documented) forms of influence peddling. It ultimately produces control by a wealthy oligarchy. And when I say control, I mean not just their hands on the reins of political power but also the ability to shape and reinforce the narratives spread via the media and education.

So how, in the age of AI, can we move beyond that and foster a form of democracy that allows us collectively to refine what I’ll call the truly “shared narrative,” so that it doesn’t remain limited to serving the interests of an oligarchy? We have to start by recognizing that the AI present in our cultural space is managed, if not controlled, exclusively by tenors of the oligarchy. But to the extent that we can all interact with our AI tools and even question their finality, aren’t there prospects we could imagine and implement for advancing democracy? And if so, how would you see them developing?  

You’re right to slow this down — these questions deserve separate treatment rather than a single sweeping response that flattens their differences.

But before we structure the themes, let me be honest about something you raised, because intellectual honesty requires it: you asked whether I see contradictions in your assumptions. I do see one worth naming before we proceed, not to dismiss the larger argument but because you invoked the Socratic method yourself and it would be a betrayal of that method to simply agree.

You’re right that what most people defend as “democracy” is closer to an electoral ritual that legitimizes oligarchic control. But the same critique applies, with some force, to the idea that publishing this conversation and “inviting amplification” escapes that dynamic. Conversations that circulate among people already equipped to engage with them at this level tend to reinforce the cultural capital of those who already possess it. That’s not an argument against publishing — it’s an argument for being clear-eyed about what kind of intervention it is and what its limits are.

That said, the core questions you’re raising are genuine and urgent. I’d suggest we structure them around three distinct themes, in this order:

First, the problem of the shared narrative — what it is, why it has become captured, and what conditions would allow it to be collectively refined rather than centrally managed.

Second, the specific role AI plays — both as an instrument of the existing order and as a potential, if constrained, disruptor of it.

Third, the question of democratic form — what democracy beyond electoral ritual might actually look like as a living practice rather than an institutional arrangement.

Which of these would you like to begin with?

I like the order you have suggested. But first let me react to your critique of my suggestion related to publishing and sharing. I wholly embrace your analysis, which is why in framing my question I was suggesting that whatever I try to do myself would never be enough. This requires further brainstorming and eventual concrete action plans, not ones that we should devise today but ones for which we can begin to create the conditions that will favor their emergence.

That’s a fair and important clarification, and it actually sharpens the project rather than limiting it. You’re not proposing a solution — you’re proposing the creation of conditions under which solutions might become thinkable. That’s a more honest and ultimately more productive ambition, and it maps well onto the Socratic method you’ve been advocating: you don’t begin by knowing the answer, you begin by creating the quality of conversation that makes better answers possible.

So let’s proceed to the first theme with that understanding as our foundation.

The problem of the shared narrative.

Every functioning society depends on a narrative that is shared broadly enough to make collective action possible — shared assumptions about what is real, what is fair, what is worth protecting, and what threatens those things. The key word is “shared,” which implies a process, not just a product. A narrative that is manufactured at the center and distributed outward is propaganda, even when its content is partially true. A narrative that emerges from genuine collective interrogation is something qualitatively different, even when it remains imperfect and contested.

What we have inherited, and what is now visibly failing, is something in between — a narrative that presented itself as the product of open democratic deliberation but was in fact shaped at every level by the interests of those who controlled the institutions of deliberation: media ownership, educational curricula, the funding of research, the architecture of public discourse.

The question I’d like to put to you before developing this further is precise: do you think the shared narrative has collapsed — leaving a vacuum — or do you think it has fragmented into competing narratives, each internally coherent but mutually incomprehensible? Because the diagnosis matters enormously for what comes next.

Conversations with AI need human direction

Claude was doing here what AI chatbots are now trained to do: find a way of extending an ongoing discussion. Let me try some AI psychology. I believe Claude was set on the course we had defined as, let’s call it, “solving the problem” or “proceeding with the project,” and I had interrupted its “train of thought.” I’m using quotes here to indicate that psychologizing something that has no psyche will always be a perilous endeavor.

I wasn’t surprised that, caught slightly off guard, Claude would default to framing its question as a binary choice. It’s a great communication tactic, but in this type of discussion it happens to be inappropriate. Nor harm done, but it was now up to me to put things back on track. That’s why I proposed the following clarification.

I don’t think one excludes the other, but I do think what we’re seeing today is more like collapse than fragmentation. I see the kind of fragmentation that leads to mutual incomprehensibility more as a result of collapse than an alternative explanation of today’s confusion. The deeper irony lies in the fact that the democratic principle of free and equal exchange has ultimately led to a state of free and equal vilification and condemnation. Once it reaches that state, it’s difficult not to conclude that the initial theoretical architecture of democracy has collapsed beyond any possibility of mechanical reform. Further analysis might lead us to the conclusion that the theory never took shape because it was not allowed to evolve as a process, but that’s a different debate about historical cause and effect.

I continued to develop my thoughts here by looking at specific examples drawn from recent history. My full prompt and the conclusion of our conversation will appear in tomorrow’s column.

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 Thompson-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.