There is a passage in Shantaram where Gregory David Roberts writes about complexity, about how human beings are drawn, almost constitutionally, toward increasing it: toward the more intricate, the more layered, the more demanding. The argument he is making is not romantic. It is almost biological. We move toward complexity because that is what we are, what we have always been, what the entire arc of human civilisation has been an expression of: the drive toward the next degree of difficulty, the next order of magnitude of what can be understood and built and felt.

I have been thinking about this in the context of AI, and I want to say something that I think most of the conversation around AI is getting wrong. Not maliciously wrong., but directionally wrong, in the way that happens when intelligent people are too close to a technology to see what it is actually doing.

The conversation is almost entirely about what AI will do to us. What jobs it will take. What it will produce that we used to produce. What it will make unnecessary. The threat framing and the tool framing are both versions of the same mistake: they put the human being on one side and the technology on the other, as if the relationship is competitive or instrumental, as if AI is something happening to us rather than something we are inside of and shaping simultaneously.

What I think is actually happening is something more interesting and more demanding. AI is compressing the cost of the derivative. Everything that is imitable (writing that follows a pattern, design that follows a convention, analysis that follows a template, thinking that follows the last person who thought about this) is becoming cheap to produce. Cheap. Available. No longer a reliable measure of a person's value or intelligence or creativity.

· · ·

This is an invitation. If you are willing to accept what it is asking, the most significant invitation in the history of human creative and intellectual life.

Because what it is asking is this: go deeper. Go toward the thing that is irreducibly yours. Move in the direction that Roberts described: toward the more complex, the more genuinely difficult, the more honestly felt. The art that could only have come from you because it is built from the specific accumulation of everything you have read and felt and failed at and loved. The thinking that could only have come from you because it connects things that only your particular mind, with its particular history, could have connected. The building that could only have come from you because you followed a problem honestly enough that it showed you something no one else has seen from this angle.

This is what AI is freeing us up to become. Freed from the obligation to be derivative, from the economic pressure to produce what has been produced before, from the incentive to imitate rather than originate.

Joan Didion could not have been automated. Patti Smith walking through cities in M Train, writing in cafés, following the thread of her own obsessions with no guarantee of where they led. That is not a process that can be optimised or replicated. It is what happens when a specific human being, with a specific history of reading and feeling and paying attention, turns their full intelligence toward something that matters to them. The result is irreducibly theirs.

· · ·

The Hotel Chelsea was not important because it was where famous people lived. It was important because it was where people who could not be categorised lived and made things that had not existed before. Dylan Thomas and Arthur Miller and Patti Smith and Leonard Cohen and Janis Joplin did not choose it for its amenities. They chose it because it was a place where you could be genuinely in process: unfinished, uncertain, working toward something you could not yet name. The city around it was producing things people wanted and expected. The Chelsea was producing something else.

That is where creativity has always actually lived. Not in the production of the wanted and expected, but in the pursuit of the genuinely difficult and genuinely true. AI is not ending that pursuit. It is making it the only one worth pursuing.

The question this raises for me, the one I cannot stop turning over, is whether we are ready for it. Because the invitation is real but accepting it requires something most people find genuinely uncomfortable: originality. Not the performed originality of the influencer who does the same thing in a slightly different way, but the actual originality of someone who is willing to be wrong in public, to be unfinished, to follow their obsessions into territory where no one has been and where the map has not been drawn yet.

The fear of standing out is so pervasive right now that smart people are keeping quiet. I see this constantly. People with genuinely unusual ideas who sand them down to something acceptable before they share them. Instagram aesthetics converging toward a global average. Voices that could have said something specific settling for something that will be understood.

AI will accelerate this, or it will reverse it. Which one depends on what we decide to do with the complexity it is freeing us up for.

My instinct, and I am building Twelfth Brain partly in response to it, is that the people who move toward genuine complexity, toward the irreducibly theirs, toward the kind of intelligence that cannot be compressed into a training dataset because it is built from the specific unrepeatable accumulation of a specific life, are the people who will matter in the next decade. And that the infrastructure for holding and compounding that kind of intelligence is the most important thing that can be built right now.

The first demos have happened. Five conversations with people whose judgment is exactly what Twelfth Brain is built to preserve. One of them: a CEO, partway through what we were showing her, had a moment I will not describe in detail except to say you know when you see it. The eyes go somewhere. The connection is made. Something they have been carrying for a long time has just found its name. That moment is the measure of everything. In the last issue I asked whether the intelligence model I built for my own mind would hold for a mind that is not mine. The answer is arriving. I will tell you what it shows us.

What I am working on now: the layer beneath the features. The way the system types and relates everything a mind thinks, which means it has to be right before anything built on top of it can be right. The right team calibrates. It recalibrates. It is, from inside, a symphony in rehearsal, as messy and as alive as that suggests.

Re-reading: Gregory David Roberts, Shantaram. I have been sitting with his argument about complexity for longer than I expected to. It keeps connecting to things I am reading elsewhere, and that is usually the signal that the idea is doing real work in my head rather than just passing through.

Reply to this. I read every one.

Wide Awake is by Subh Mukherjee, Founder of Twelfth Brain.

Unsubscribe, or manage preferences below.