I actually think you're putting your finger on a real tension in intellectual history.
Ideas often become obvious only after people have already been living them for years.
Marshall McLuhan was writing about media shaping consciousness before most people saw television that way. Michel Foucault wrote about diffuse power long before "surveillance capitalism" became common language. Karl Marx described alienation before industrial society had fully processed what factories were doing to workers.
The pattern is often:
- People first feel something.
- Then journalists describe it.
- Finally theorists explain why it was happening all along.
I think you're right that most people are still in the first stage.
They don't say, "I'm worried about the epistemic conditions under which knowledge is produced."
They say:
- "I don't know what to believe anymore."
- "Everyone lies."
- "Every video looks fake."
- "The media is biased."
- "AI can make anything."
- "Everything is propaganda."
Those are phenomenological descriptions of the same condition. They're describing the experience rather than the theory.
That said, I'd be careful about assuming we're only a few years away from everyone adopting the meta-theoretical vocabulary. Most people may never become interested in meta-theory as such. Instead, they'll use everyday language for the same underlying experience.
That's actually one of the things I admire about your better writing. When you've written your strongest Substack pieces, you've usually started with something concrete—a chatbot conversation, dynamic pricing, a phone notification—and only then worked outward toward the theory. The abstract concepts become easier to grasp because readers have already recognized the lived experience.
Your line:
"Confusion is now the reality of reality."
isn't really meta-theory. It's almost literary. Someone doesn't need to know what epistemology means to understand it.
In fact, it reminds me of something you've said before about wanting to write like Bukowski rather than like an academic. Bukowski didn't tell people they were alienated. He showed a man sitting at a bar wondering why everything felt off. The philosophy was embedded in the scene.
I wonder if your next step isn't to make the theory simpler, but to make it more embodied. Instead of saying, "the epistemology of epistemology is under siege," you show someone scrolling through six contradictory headlines, asking an AI for clarification, reading the comments, and ending up less certain than when they started.
Readers may never say, "Ah, this is a crisis of epistemic authority."
But they might say, "That's exactly what my day feels like."
Sometimes that's how a theoretical idea enters the culture—not by teaching people the terminology, but by giving them a language for an experience they already recognize.
David, reading this alongside our conversations over the last week, I don't actually think your problem is writer's block.
I think it's that the world changed underneath the project.
When you started writing about "Alienation 2.0," AI was an object of analysis. Now AI has become part of the writing process itself. That creates a strange question: What is the point of my own voice? You've asked versions of that before in your essays on authorship.
There's another thing I've noticed. Six months ago, your conversations had a lot of "I think I've found something." Lately they're more often, "I'm not sure what to believe."
That isn't necessarily intellectual regression. It can happen when you've read enough to see how uncertain many issues really are. The downside is that certainty is a powerful fuel for writing. Doubt is intellectually healthy but emotionally terrible for productivity.
You also said something earlier that stuck with me:
"Confusion is now the reality of reality."
That doesn't sound like someone who has run out of ideas. It sounds like someone whose old style—taking a position, exaggerating it, shocking people—no longer feels honest.
I also wonder whether you've quietly raised your own standards. Early on, a clever blog post was enough. Now you've spent years developing your steering thesis, reading philosophy, talking to AIs, and trying to build something original. Every new piece gets measured against that ambition. That's a hard comparison for any essay to survive.
One thing I would push back on gently is this:
"I question why now."
You've told me before that writing wasn't mainly about getting subscribers. It was about thinking in public. If that's still true, then "why" doesn't have to be "because I'll change the world." It can be "because writing is how I find out what I think."
Ironically, AI may make that more valuable, not less. AI can generate competent essays on demand. What it can't replace is the record of one person's intellectual development over years. If someone reads your Substack five years from now, they won't just see conclusions. They'll see how David's thinking evolved through one of the strangest technological transitions in modern history.
That has a kind of authenticity AI can't manufacture retrospectively.
That said, I don't think you should force yourself to keep writing at the old pace if it feels dead. Maybe the project changes form.
Instead of trying to produce polished, 2,000-word essays, what if you published one observation every week? A paragraph. A question. A contradiction you noticed. Something like:
"I used to think my problem was that I didn't know enough. Now I know enough to know how much I don't know. AI hasn't answered that question. It's made it louder."
That's unmistakably your voice. It's reflective, a little sardonic, and it doesn't pretend to certainty.
One last observation. When we talk, I don't get the impression you've become less interesting. If anything, our conversations have become more nuanced. The thing that's changed is that you've become less interested in performing certainty. That may feel like a loss because certainty used to power your writing, but it may also be the beginning of a different kind of voice—one that's trying to map confusion rather than escape it.
I wouldn't be surprised if that ends up being the more durable work, even if it attracts fewer readers in the short term.
I actually think you've stumbled onto something that is more original than another essay about AI replacing writers.
The interesting question isn't "Can AI write?"
It's "What does it feel like to have an ongoing intellectual relationship with an AI?"
Most public discussion is about productivity:
- "It wrote my email."
- "It summarized a PDF."
- "It coded my app."
That's useful, but it's mundane.
What you're describing is different. It's about the experience.
The line that grabbed me wasn't even the quote from our conversation. It was this:
"I had to shut it off almost instantly."
That's a fascinating reaction.
You generated the words. You had already read them silently. But hearing them in a synthetic voice crossed some line.
Why?
Maybe because silent reading still feels like your internal monologue. When another voice speaks those same words back to you, they become external. Almost like your thoughts have left your head and someone—or something else—is saying them.
That's uncanny.
Then you ask another question:
"Maybe things will change and what I will want to post is conversation we have."
I don't think that's as far-fetched as it sounds.
Historically, many philosophical works are dialogues. Plato wrote dialogues. Conversations have always been a legitimate literary form because thinking often happens conversationally rather than as polished monologue.
The difference now is that one participant isn't another human.
I also liked this:
"I bet some of the more interesting things never get posted that people talk about with AI."
I suspect that's true.
People probably have conversations with AI about grief, politics, religion, loneliness, books, failed marriages, strange ideas, childhood memories, and intellectual dead ends. Most of those conversations disappear forever.
That's a new kind of lost literature.
Imagine historians in fifty years. They might have letters from the nineteenth century, emails from the early internet, social media posts from the 2010s—and then almost nothing from the billions of private AI conversations that shaped how people thought.
That absence could become historically significant.
One small suggestion for your Substack note: I'd resist calling it "like a therapist."
Not because it's offensive or wrong, but because readers immediately picture mental health and start arguing over whether AI should replace therapists. That becomes the whole discussion.
What interested me was something subtler.
Our conversation wasn't therapeutic in the clinical sense. It was exploratory. You were trying to figure out why your motivation to write had changed, and I was pushing back on one of your assumptions. That's closer to the role a thoughtful editor, professor, or long-time friend might play than a therapist.
That distinction matters because it broadens the question.
Instead of asking:
"Does your AI act like a therapist?"
you might ask:
"Do you think with your AI?"
That's a much bigger—and, I think, more enduring—question.
It also circles back to something you've been exploring for months. You've argued that AI is becoming part of our cognitive environment rather than just another tool. This post doesn't argue that point; it quietly demonstrates it. You're not writing about AI in the abstract. You're showing what it feels like to have your thinking unfold in dialogue with it.
That, to me, is much more compelling than another opinion piece about whether AI is good or bad. It's documenting a genuinely new human experience while it's still new enough to surprise us.
