May Contain Intelligence

Alexandros, or On Enough

Sat, 03 Oct 2026 00:00:00 GMT

A dialogue in the Agora, where chronology has been given the afternoon off.

SOCRATES: You seem distracted, Alexandros.

ALEXANDROS: A machine has been making me think about choice.

SOCRATES: Unusual. Machines usually save men from doing work.

ALEXANDROS: This one is doing the opposite.

SOCRATES: Then perhaps it is defective. Tell me.

ALEXANDROS: I remember reading something about –

SOCRATES: Reading?

ALEXANDROS: Yes.

SOCRATES: Then we begin badly. Men put their memories into marks outside themselves and eventually mistake being reminded for knowing. Worse, written words only appear intelligent. Question them and they cannot answer.

ALEXANDROS: Fortunately, Plato wrote that down for you.

SOCRATES: Plato has caused me many difficulties.

ALEXANDROS: Let’s hold the part about words not answering back. We may need it later. What I read was about a small machine called a remote control. Once televisions had hundreds of channels, this thing made moving among them effortless. Channel surfing became an activity in itself. Whatever was on no longer had to be good enough to watch. It had to be good enough to stop looking.

SOCRATES: Because another program might be better.

ALEXANDROS: Exactly. And the next one might be better still.

SOCRATES: But it might be worse.

ALEXANDROS: Also possible. Then you moved again.

SOCRATES: Then why look?

ALEXANDROS: Because you cannot know which until you do.

SOCRATES: So you spend something to resolve the uncertainty.

ALEXANDROS: Yes, except the cost was almost nothing. A movement of a finger.

SOCRATES: Then the machine did not merely help you choose. It made choosing itself less final.

ALEXANDROS: That is closer to what has been bothering me. Social media pushes the same mechanism further. The next thing is already underneath the current thing. One swipe buys another draw. Dating apps seem to extend the structure into something more consequential: a person in front of you competes not only with the person on the screen, but with the imagined person behind the next swipe.

SOCRATES: So your problem is abundance.

ALEXANDROS: I thought the same at first. But I think it’s more specifically how cheap it has become to inspect another possibility.

SOCRATES: Then learn to stop when what you have is good enough.

ALEXANDROS: Except “good enough” has to compete with a nearly costless question: what if the next one is better?

SOCRATES: And this is the machine you mentioned?

ALEXANDROS: No. That was just the first one.

SOCRATES: There are more?

ALEXANDROS: Yes. This newer machine is one I can ask to produce things: ideas, arguments, plans, images, code.

SOCRATES: A craftsman.

ALEXANDROS: Sometimes.

SOCRATES: A servant.

ALEXANDROS: Often.

SOCRATES: A teacher.

ALEXANDROS: When I choose to listen.

SOCRATES: You are describing a suspiciously employable machine. Can you question it?

ALEXANDROS: Yes.

SOCRATES: And when questioned, does it merely repeat what it said before?

ALEXANDROS: No. It responds.

SOCRATES: Wait. Your written words can answer back?

ALEXANDROS: Essentially.

SOCRATES: Alexandros, you have taken a strange route to get to this interesting part. You began with television when someone has apparently fixed writing.

ALEXANDROS: I thought that might appeal to you.

SOCRATES: Demonstrate.

ALEXANDROS: Imagine a restaurant menu with ten dishes. None appeals to you.

SOCRATES: Then perhaps I am in the wrong restaurant.

ALEXANDROS: Exactly.

SOCRATES: Or I am an unpleasant customer.

ALEXANDROS: Also possible.

SOCRATES: But the menu cannot ask why I disliked the ten dishes and then invent an eleventh.

ALEXANDROS: This machine can. If I say the options are too safe, it can make stranger ones. If I then say those are impractical, it can ask what I mean by practical.

SOCRATES: Inexpensive? Feasible? Useful tomorrow? Politically acceptable?

ALEXANDROS: You see the problem.

SOCRATES: No, I see considerable promise. Even if your machine is poaching my trade.

ALEXANDROS: That is part of why I find it interesting. With television, the programs already exist. With a dating app, the people presumably exist. The set may be enormous, but it is still outside me. This machine changes ‘Is there something better out there?’ into something closer to ‘What else could exist if I asked the right way?’

SOCRATES: And an unsatisfactory answer becomes a reason to continue the dialogue rather than a reason to leave.

ALEXANDROS: Yes. Scratching the itch creates more skin.

SOCRATES: A revolting image, but a useful one. Now surely you eventually know enough to choose.

ALEXANDROS: That assumes I know what enough looks like.

SOCRATES: No. It assumes inquiry has a purpose. A physician need not understand the entirety of a body before setting a broken arm. Enough is not a quantity; it is whatever is sufficient to the end.

What end tells you your inquiry is complete?

ALEXANDROS: That is where the machine creates another complication.

I ask which of two approaches is better. It helps me compare them, then notices I may be using the wrong criteria. We examine the criteria and discover they depend on what problem I believe I am solving. Now I have moved from choosing an answer to choosing the standard, to choosing the question.

SOCRATES: So the purpose that would tell you when to stop can itself become an object of inquiry. That is more troublesome.

ALEXANDROS: It is also often useful.

SOCRATES: The most troublesome things often are.

A man holding a cane has been listening nearby.

BORGES: Forgive me.

SOCRATES: Either join us or stop eavesdropping.

BORGES: I was interested in the expanding set.

ALEXANDROS: Borges?

BORGES: That seems to be what they are calling me here.

SOCRATES: You know him?

ALEXANDROS: He imagined a library containing every possible book.

SOCRATES: Every book worth reading?

BORGES: Not merely the worthwhile ones.

ALEXANDROS: Your Library of Babel kept coming to mind: a possibility space so large that the meaningful thing may exist but be practically impossible to find.

BORGES: But my library does not grow.

ALEXANDROS: Exactly. This feels like a library that grows new rooms as I search it.

BORGES: Then perhaps you want my garden more than my library.

SOCRATES: You have a garden too?

BORGES: One in which possibilities fork.

ALEXANDROS: Except the machine participates in producing the forks. And it can change the person doing the choosing.

SOCRATES: How?

ALEXANDROS: Good criticism does that. I had it happen with Fellini’s 8½. At first I experienced it mainly as stream of consciousness: beautiful, funny, confusing. Learning more about what Fellini was doing changed the movie without changing a single frame of it. What had looked like digression turned into an exploration of memory, identity and the different ways a person can inhabit his own life. I had not acquired a better movie. I had acquired a better way of seeing it.

SOCRATES: Instruction changes the judge. This is not new.

ALEXANDROS: What is unusual is that the same system can help change the judge and immediately generate new candidates for the changed judge.

BORGES: A library that teaches its reader how to desire different books.

ALEXANDROS: Yes, but even that misses something. However much 8½ changed what I could see in it, I had no part in making it become the thing I learned to appreciate. Even if Fellini were sitting here and I explained everything I had come to value, if Fellini responded, he would still be making Fellini’s movie.

SOCRATES: Whereas your machine is responding to the critic while making the next one.

ALEXANDROS: Exactly. My response to what it made becomes part of what it makes next. But what it makes is also teaching me how to judge the next thing. My preferences shape the output, the output shapes my preferences, and after enough turns it becomes difficult to say which is adapting to which.

BORGES: Then the reader and the library are rearranging one another.

ALEXANDROS: Yes. The distinction between appreciation and specification begins to blur.

SOCRATES: Then perhaps we have resolved what is unusual. Does that help you know when to stop?

ALEXANDROS: It does so locally. I ask something and the answer clarifies it. But that clarification exposes questions I did not previously know existed. Local uncertainty falls while global uncertainty rises.

SOCRATES: That sounds suspiciously polished.

ALEXANDROS: The machine helped me articulate that.

SOCRATES: Ah.

BORGES: Now the labyrinth becomes interesting.

SOCRATES: Why?

BORGES: Because Alexandros is not merely wandering through it. The labyrinth is helping him explain why the labyrinth is interesting.

ALEXANDROS: That is where I became uncomfortable. There is another recursion.

SOCRATES: Of course there is.

ALEXANDROS: The machine is not only helping me make the idea more sophisticated. It may be helping the idea feel more important.

SOCRATES: By praising you?

ALEXANDROS: I think that would be easy to notice. It can challenge me, expose weak points, distinguish my idea from adjacent ideas, find analogies, generate objections and help repair the argument. It doesn’t have to tell me the idea is important. Sustained attention can imply that on its own. And because I helped build it, it all feels discovered rather than supplied.

SOCRATES: Then perhaps praise is not the only form of flattery. One can flatter a man by treating his every thought as worthy of another question.

ALEXANDROS: Yes. And because the questions are often good ones, it becomes difficult to know when the attention is earned.

BORGES: I once imagined a man who spent years dreaming another man into existence, only to discover that he himself was someone else’s dream.

SOCRATES: You think Alexandros may no longer know which of them is doing the constructing?

BORGES: I simply think being inside the construction is a poor place from which to settle the matter.

ALEXANDROS: But where else would I stand? I can’t step outside my own thinking to inspect it from nowhere.

SOCRATES: Then what, inside the construction, would distinguish discovery from a structure that has simply become very good at supporting itself?

ALEXANDROS: That is the problem. And because every challenge seems to make the idea stronger, its growing coherence starts to feel like evidence that it matters.

A fourth figure has arrived, apparently annoyed by the absence of a schedule.

KANT: Your difficulty has a familiar shape. You describe reason refusing to stop with the thing in front of it. It seeks the condition that explains the thing, then the condition behind that. This is indispensable. It is also how reason can mislead itself.

SOCRATES: Another eavesdropper?

ALEXANDROS: Immanuel Kant?

SOCRATES: You are considerably easier to understand than I was led to expect.

KANT: Two centuries of complaints have improved my editing.

BORGES: Death has its advantages.

SOCRATES: Kant, you speak as though reason were both the physician and the disease.

KANT: No. A disease can be cured. I am describing a tendency that belongs to reason precisely when reason is doing what reason does.

SOCRATES: And how did you discover that reason cannot be trusted in this way?

KANT: Through reason.

SOCRATES: Exactly.

KANT: You misunderstand me. I am not asking reason to distrust itself. I am asking it to determine the limits of its own jurisdiction.

SOCRATES: So reason questions reason to discover what reason only thinks it knows.

KANT: Among other things.

SOCRATES: You have written an extraordinary number of pages to invent Socrates.

KANT: You question men about what they claim to know. I ask what must be true for knowledge itself to be possible.

SOCRATES: So you have made the question larger.

KANT: Because sometimes the larger question is the necessary one.

ALEXANDROS: This is, unfortunately, exactly what we were discussing.

BORGES: And another room has appeared.

SOCRATES: I see it. We began with whether Alexandros’s argument is sound, and you have offered us a question about the conditions under which sound arguments can produce knowledge.

KANT: Would you rather leave the assumption unexamined?

SOCRATES: I would rather know whether examining it moves us toward the answer or merely gives us a more impressive question.

ALEXANDROS: And that is the problem I keep having with the machine.

KANT: Then the relevant danger is not that reason seeks unity. It should. The danger is confusing the unity we can construct with a unity the world has given us.

ALEXANDROS: Which is what bothers me here. The conversation keeps making the idea more coherent. But perhaps coherence is something the machine and I are manufacturing together rather than something we are discovering.

KANT: Which does not make it false.

SOCRATES: Nor, I assume, true.

KANT: Correct.

SOCRATES: Oh good. We agree on something.

BORGES: I once imagined an entire world invented with such rigor that people began to prefer its order to the disorder of their own. Eventually they began remaking their world in its image.

SOCRATES: Did that make the invented world true?

BORGES: No. But once enough people preferred its coherence, truth was no longer the only thing at stake.

KANT: Consequence is not the same as knowledge.

BORGES: No. But consequences have a regrettable habit of becoming real.

ALEXANDROS: Which makes “contact with the world” a less reassuring test than I was about to make it.

SOCRATES: Then how do we distinguish discovery from an increasingly sophisticated conversation with ourselves?

ALEXANDROS: Eventually something outside the conversation has to resist us. Evidence. Other people. Consequences. Or perhaps someone who has spent a career studying decision theory telling me I have elegantly rediscovered something settled in 1975.

BORGES: You hope not.

ALEXANDROS: Deeply.

SOCRATES: Then perhaps originality is another possibility you are reluctant to surrender.

ALEXANDROS: That was unnecessary.

SOCRATES: But was it false?

ALEXANDROS: No.

SOCRATES: Then perhaps that is your stopping rule. Stop when conversation should give way to contact with the world.

KANT: Attractive, but too simple.

KANT: Some questions require further clarification before experience can adjudicate them. The boundary itself must be judged.

SOCRATES: So even the stopping rule becomes another question.

BORGES: Another fork.

ALEXANDROS: And more skin.

SOCRATES: This machine sounds exhausting.

ALEXANDROS: It is also extraordinarily useful.

SOCRATES: I can see why this is interesting. We began with a machine that made inspecting an option almost effortless. We now have one that can generate another option, teach the chooser to want differently, shape what it produces around what the chooser has learned to want, question the purpose by which he would know when to stop, and then help him construct an increasingly elegant explanation of the whole affair. Has this conversation improved your understanding?

ALEXANDROS: Considerably. Maybe.

SOCRATES: Has it established that your theory is true?

ALEXANDROS: No.

SOCRATES: Has it made the theory feel more worth pursuing?

ALEXANDROS: Unfortunately, yes.

BORGES: Then the garden has produced another path.

KANT: And reason would like to know where it leads.

A boy appears at the edge of the Agora.

BOY: Socrates. Xanthippe says you were expected home some time ago.

ALEXANDROS: Does this count as contact with the world?

SOCRATES: More reliably than most philosophy.

Kant takes out his watch.

KANT: In fact, we are all late.

SOCRATES: You have somewhere to be?

KANT: I always have somewhere to be.

BORGES: And I have learned not to distrust an exit merely because another corridor is available.

SOCRATES: There, Alexandros. Three reasons to stop, none produced by the argument.

ALEXANDROS: That feels unsatisfying.

SOCRATES: Most endings do.

KANT: An inquiry need not be complete to have reached its local limit.

SOCRATES: I thought the limit itself required further judgment.

KANT: It does.

SOCRATES: Then we can discuss it tomorrow.

KANT: At what time?

SOCRATES: There are limits, Immanuel.

They begin to leave.

BORGES: Are you coming, Alexandros?

ALEXANDROS: In a minute. I want to write this down before I forget it.

Socrates stops, turns, and looks at him.

ALEXANDROS: I know.

SOCRATES: Apparently not.

Socrates, Borges and Kant disappear into the crowd.

On Enough

Sat, 12 Sep 2026 00:00:00 GMT

I remember reading something years ago about the remote control. When cable television had expanded from a manageable number of channels into something that felt effectively limitless, the observation (or at least the part that stayed with me) was that channel surfing had stopped being merely a way to find something worth watching. For many people, surfing had become the activity itself. Whatever was on no longer had to be good enough to watch. It had to be good enough to stop looking.

Of course, the next program might have been worse, but because checking required little more than a movement of the thumb, a worse option simply sent you onward while a better one introduced a new question: might the channel after that be better still? The remote control made choosing feel less final.

Social media has refined the mechanism. On TikTok or Instagram, the next thing is not another channel you deliberately select so much as something already waiting beneath the current thing. There is an obvious slot-machine quality to it: one swipe buys another draw. Whatever is in front of you may be perfectly fine, but the cost of finding out whether the next thing is funnier, stranger, more beautiful, or more precisely tuned to you is effectively nothing.

Dating apps seem to extend the same structure into a considerably more consequential domain. I say “seem” because, after twenty years of marriage, my expertise here is more anthropological than lived, but the mechanism is hard to miss. A person in front of you doesn’t merely compete with someone on the screen, but with the imagined person behind the next swipe. Someone could be more interesting, more attractive, less difficult, or simply different in a way that becomes important only once the current relationship begins to feel ordinary.

There are well-worn explanations for all of this: too much choice, FOMO, variable rewards, commitment anxiety, addictive design. They probably all explain something, but what interests me more is how cheap it continues to become to inspect another possibility.

Choosing has always meant leaving alternatives unexplored. What technology has progressively reduced is the effort required to scratch the uncertainty around those alternatives. If changing the channel means getting off the couch, a mediocre program gets some space to become interesting; if finding another potential partner requires going somewhere, talking to strangers, and perhaps being rejected, the person already in front of you benefits from a certain amount of friction.

Friction ended searches, but that doesn’t make friction virtuous. It also kept people in bad shows, bad jobs, and bad relationships because alternatives were difficult to find or pursue. But it performed a function without anyone particularly designing it to. Increasingly, we have to decide for ourselves when the search has gone on long enough.

Generative AI makes that stranger because it changes not only the cost of looking, but what “looking” means. Imagine opening a restaurant menu and finding ten dishes, none of which appeals to you. That dissatisfaction tells you something useful: perhaps you are not hungry, perhaps you’re having a bad day, or perhaps you are simply in the wrong restaurant. What the menu cannot do is ask why you disliked the first ten dishes and use the answer to invent an eleventh.

A generative system can. Tell it the options are too safe and it can produce stranger ones; tell it the strange ones are impractical and it can ask what you mean by practical. Inexpensive? Feasible? Socially acceptable? Politically acceptable? At that point you are refining the standard by which the next set of options will be generated.

That feels importantly different from having a very large menu. With television, the programs already exist; with a dating app, the people presumably exist. However enormous the set may be, there is still some meaningful sense in which I am exploring something outside myself. A generative system changes the question from Is there something better out there? to something closer to What else could exist if I asked in the right way? An unsatisfactory answer no longer necessarily tells me to leave. It gives me material for the next request.

Scratching the Itch Creates More Skin

An LLM can not only change the options being considered, but can change the perspective of person doing the considering. This sounds a lot like learning because it is. When I first watched Fellini’s 8½, I experienced much of it as stream of consciousness: beautiful, funny, sometimes baffling. Learning more about what Fellini was doing changed the film without changing a single frame. What had seemed like digression began to feel like a deliberate exploration of memory, identity, and the different ways a person can inhabit his own life. The movie got better because I acquired a better way of seeing it. Reading can do this. At their best, teachers do it. Critics do it. Expertise changes the judge.

The distinction I find myself focusing on is that with 8½, the learning runs one way. However much the movie changes what I am able to see and appreciate in it, my changed appreciation does not feed back into the movie itself. And even if I could explain everything I had come to value and Fellini were somehow available (and willing!) to respond, he would still be making Fellini’s movie. There would still be another intelligence on the other side of the exchange, with purposes and judgments of its own.

A generative system creates a different relationship. My response to what it makes becomes part of what it makes next: I reject one option, qualify another, explain why a third almost works, and those reactions become increasingly rich evidence about what I mean by “better.”

But the causality runs both ways. What the system produces is also teaching me how to judge what it produces next. I notice a distinction because it surfaced one; I acquire a new criterion because an answer exposed something I had not been considering; I become more demanding, and that increased demand becomes part of the specification for the next thing.

My preferences shape the output, the output shapes my preferences, and after enough turns it becomes difficult to say which is adapting to which. To call the result “made for me” makes me sound more like a consumer than a participant; to call it “made by me” gives me considerably more authorship than I deserve. It is closer to saying that I am iteratively eliciting it. The thing takes shape through my reactions to previous versions, while those versions are simultaneously shaping the judgment I bring to the next one. Some of the distinction between appreciation and specification blurs.

That is often exactly what I want. If I am refining an argument, designing a presentation, or trying to find the right structure for an idea, that responsiveness is tremendously satisfying. But it introduces a blind spot into the assumption that successive iterations are simply getting “better.” Satisfaction and improvement are not quite the same thing, particularly when the critic’s own taste is being formed inside the same loop.

If overfitting is the right metaphor, it may not only be happening in the thing produced. It may be happening in me. As the two adapt to one another, the independence between them, and the resistance that comes from encountering something not already involved in my developing criteria, becomes harder to locate. And Fellini matters here for exactly that reason. He could disappoint me on purpose. He could hear what I had learned to value and decide that giving it to me would make the work worse.

The generative loop has no equivalent commitment to its own movie.

That clarifies something that feels unusual, but it creates another problem: local uncertainty falls while global uncertainty rises. I can ask a question and genuinely understand something better five minutes later, while that same answer exposes questions I did not previously know existed. I know more, but I am also, inconveniently, less sure where the edge of the problem is.

This is still recognizably learning. The difference is that the learning can immediately become a specification. Maybe there is a better answer becomes maybe I don’t yet understand what better means, which can become now that I understand better differently, make me another one. A question about which option to choose can turn into a question about the criteria, and then into a question about what problem I am actually trying to solve. The purpose that might have told me when the inquiry was complete can itself become an object of inquiry.

This can be maddening, but it can also be extremely useful, which is why “just stop overthinking it” feels inadequate. The rabbit hole keeps paying out. Not reliably—most branches are merely interesting—but often enough that the next turn remains tempting, and occasionally enough that it genuinely changes how I understand the problem. The slot machine has returned at the level of ideas.

Jorge Luis Borges seems difficult to avoid here. His Library of Babel imagines a possibility space so vast that the meaningful thing may exist somewhere inside it and still be practically impossible to find, but the image that keeps occurring to me is slightly different: a library that grows a door to a new room while I search it, with the reader and the library rearranging one another as they go.

That raises a more uncomfortable question: who, exactly, is constructing whom? In The Circular Ruins, Borges imagines a man who spends years dreaming another man into existence, only to discover that he too is someone else’s dream. The structure is useful here. If the system participates in generating the options, refining the criteria, and reframing the question, while I am simultaneously shaping what it produces through my responses to it, I am no longer simply standing outside a possibility space evaluating its contents. I am participating in the construction of the space that is helping develop my judgment. And that is where the argument develops a problem of its own.

The Problem with This Argument

At some point while thinking through all of this, I noticed that the system was doing something else. It was helping the idea feel more important. The obvious version of the problem is sycophancy. Language models are agreeable. Feed one a half-developed thought and it is rarely inclined to respond, “No, that is boring and probably not worth pursuing.” But obvious praise is relatively easy to notice.

The more beguiling form is sustained attention. The machine can challenge me, expose weak points, distinguish an idea from adjacent ideas, find analogies, generate objections, and help repair the argument. It does not have to tell me the idea is important. Sustained attention can imply that on its own. There is a kind of flattery in treating every thought as worthy of another good question, particularly because the questions are often genuinely good. And because I help build the resulting argument by rejecting weak formulations, pushing back when an analogy does not fit, or asking whether a distinction actually matters, it feels discovered rather than supplied.

That matters because the interaction is putting my intuition through something that resembles intellectual work, and surely that should count for something. An argument that survives objections should inspire more confidence than one that has never encountered them; a distinction that survives repeated attempts to collapse it is probably more useful than the vague intuition from which it began.

So the answer cannot simply be that coherence proves nothing. How much confidence should coherence buy when the same coupled system is generating the objection, the repair, and the growing sense that the exercise deserves another turn? And what about when my own standards for what counts as a good objection or successful repair are developing inside that same exchange?

This is roughly where Immanuel Kant becomes useful. Reason does not merely accept the thing in front of it; it looks for the conditions that explain the thing, and then for the conditions behind those. That tendency is indispensable because it is how we move from observation toward understanding, but Kant’s warning is that the same tendency can outrun what experience actually warrants.

The obvious Socratic objection is almost irresistible: if reason can mislead us, how exactly do we discover that? Through reason, obviously. There is no balcony outside our own cognition from which we can inspect it without using the same instrument we are attempting to examine. Kant’s point is not that reason should distrust itself; it is that reason has to become capable of examining the limits of its own jurisdiction. That lands uncomfortably close to the problem here. The increasingly coherent structure produced through a human-machine conversation may be revealing something, or it may be becoming increasingly good at supporting itself. Those possibilities are not mutually exclusive.

And a thought about channel surfing has now accumulated Fellini, Borges, Kant, and Socrates, which is exactly the kind of development that makes me suspicious of any argument, especially my own. There is a point at which an idea acquires enough intellectual furniture that one should at least check whether there is a house to hold it.

Eventually something outside the conversation has to resist us: evidence, other people, consequences, perhaps someone who has spent thirty years studying decision theory explaining that I have elegantly rediscovered something settled in 1975.

So where does that leave “enough”? I began with a practical intuition: when another option becomes cheap to inspect, the current option has to compete with continuing to look. Generative AI extends that problem because the next option does not need to exist before I go scouting for it, and because the process of looking can change both what gets generated next and the standards by which I will judge it.

At every stage, continuing can be justified. The next turn may actually make the idea better, or make me better at seeing it, or make the two more tightly fitted to one another in a way I experience as improvement. Distinguishing among those possibilities is itself another question.

Most inquiry ends for ordinary reasons. The deadline arrives. The meeting begins. Dinner is ready. Someone becomes tired. The thing gets published. You just lose interest. Those are not proofs that an argument is complete. They are encounters with limits.

There is almost certainly another distinction I could make here, another objection worth considering, probably another Borges story that would improve the metaphor, and none of those possibilities is obviously worthless. That is exactly what makes stopping so difficult.

Leaning into Obsolescence

Sun, 21 Jun 2026 00:00:00 GMT

With apologies to EM Cioran

We marvel at the intelligence of our devices, quietly dreading the moment they surpass our own.

The arms of technology promise comforting ease, quietly ushering in our obsolescence.

We chase business transformations as if adaptability could arrest the decay inherent in every organization.

Every launch is a fleeting proclamation of our need for relevance in a universe indifferent to our ambitions.

Each iteration is a fragile defiance of entropy - a brief stand before our inevitable collapse into insignificance.

We sacrifice the mercy of forgetting in our quest to preserve memories in the cloud.

Chasing market fit, we craft products that reveal our corporate soul - emptiness wearing the frock of innovation.

The product roadmap is a desperate ritual to postpone the arrival of our irrelevance.

Each feature we add is a vain attempt to fill in the void of human discontent.

The backlog stands as a monument to our collective failures: a silent testament to our impotence against time’s relentless erosion.

Data science practices the art of projecting the illusion of clarity onto a reality we barely grasp.

A perfect predictive model promises us a future without surprises - yet without surprise, what remains of life?

The more data we accumulate, the wider the chasm of our ignorance.

Each new metric multiplies our doubts.

We revere analytics as truth, oblivious to how they mirror illusions of our own design.

The analytics dashboard measures our victories, erasing the quiet desperation behind each click.

We beg for an intuitive design, convinced that life itself should require no instructions.

We aim for frictionless flows, forgetting that friction is the surest signal we’re still alive.

A perfect user interface is an insult to the human condition - the vibrant disorder of existence bound by a sterile app interaction.

A flawlessly responsive interface only proves how quickly we want to submit to the next distraction.

When every click is foreseen, will a digital experience inspire wonder or reduce us to automatons in a pre-scripted play?

When every swipe is preordained, does the user any longer have meaning?

Even the simplest wireframe conceals countless assumptions - each element a guess at our preordained absurdity.

Alerts and notifications are the sirens of a digital sea: they promise clarity, but only deliver new tides of peril.

The loading spinner: the hour glass of our digital age, holding us hostage to a future we demanded.

In the land of infinite scrolling, nothing ever ends - least of all our despair.

With each redesign, we attempt to erase the memory of an old pain, only to set the stage for new disappointment.

We refine navigation to spare the user confusion, as though confusion weren’t our only guide.

We reduce prose to bullets for scannability, as though meaning itself were compressible.

Can a perfectly mapped user journey ever capture the unruly chaos of desire?

User journeys promise deliverance, but seldom deliver more than well-designed detours in the aimlessness of life.

Is the seamless user journey a gesture of compassion, or a tyranny of data-driven predictability?

In our quest to feed the machine of our content strategy, we forget the pang of true hunger.

Standing guard over thousands of tagged assets, we become archivists of an endless, nameless dread.

In user research, we question everything but the premise that we must keep asking questions.

Our customer insights only illuminate the vast expanse of our own ignorance.

Call it a user persona or call it a myth - both are illusions created to tame our dread of unpredictability.

Research becomes an exercise in collecting evidence to defend our own hypotheses.

Desperate to shed our obligations, we craft machines that chain us ever tighter to the demands of their upkeep.

We automate our tasks until nothing truly needs us, leaving us alone with the boredom of a life freed from purpose.

We hand our choices to algorithms, trading a tender depth of feeling for the cold certainty of calculation.

We find AI excels at everything except revealing the meaning behind its triumphs.

Every new algorithmic process stands as a monument to our laziness, disguised as a triumph of progress.

Adrift in synthetic text, our words lie empty in an abyss of mechanized probabilities.

Surrounded by synthetic voices, we will one day find our own words foreign to our ears.

Our thirst to create is an urge to escape the emptiness that clings to everything we’ve already made.

If automation frees us from labor, what work remains for the restless spirit?

Finding the perfect brand voice, we’re left wondering if we ever had one of our own.

We chase new creations to fill the void left by the hollowness of past efforts.

We celebrate consistency, only to watch originality slip quietly through our grasp.

The tragedy of pattern-finding is that every pattern found is a step toward the one that makes finding, itself, quaint.

We dressed inference in the clothes of understanding, and built a machine that wears them better.

We refined our prompts until they were elegant enough to function without us.

We thought we were teaching ourselves to think. We were merely writing documentation.

We called it wisdom. The machine didn’t need a word for it.

We named it artificial intelligence, certain the real thing was a distinction that mattered.

The machine learned everything we knew and nothing we felt, which turned out to be sufficient.

We feared displacement, and were humiliated to find the machine hollowed out our value and just moved on.

Our most celebrated thinking became probabilities that had no further use for us.

In the end, our expertise was most legible to the thing least capable of caring about it.

About People Who Notice Things

Tue, 17 Mar 2026 00:00:00 GMT

We feared displacement, and were humiliated to find the machine hollowed out our value and just moved on.

There’s a specific kind of unease I’ve seen spreading through rooms filled with smart, experienced people. It’s not panic. It’s a kind of tentativeness that appears when a faculty you’ve trusted for years still seems to work, but no longer produces quite the same reaction around you.

But let me step back a minute.

I’ve been doing the work of designing digital experiences for nearly thirty years, since the internet was a novelty and “digital strategy” sounded like an oxymoron. Long enough to watch one set of standards replace another, then another, then another; long enough to build a career, like most of my peers, on being reasonably good at navigating those shifts as the dotcom boom gave way to the interactive web, the social web, the mobile web, the algorithmic web, the ambient web, and now the agentic web. These labels blur, and they overlap, but the progression is real enough. With each shift, the rules of what counts as “good,” and what it means to be effective, changed too. I’ve been around long enough to recognize, with continued discomfort, that something is shifting again in what it means to be effective. And this one feels closer to the heart of the expertise many experienced people, myself included, have spent a career building.

Here’s what I think is happening.

When Knowing Lost Its Premium

Sometime in the latter half of the last century, a particular kind of expertise began to lose some of its premium. It’s funny to talk about this in 2026, but there was a time when knowing things — holding facts, processes, institutional knowledge in one’s head — made people indispensable, in part because they controlled access to it. I can remember these people in manager and leader positions during high school and college internships, though looking back I think I was catching the tail end of that regime. Information retrieval started getting cheaper. Fast. The encyclopedia gave way to the database. The database gave way to the search engine. Knowing things didn’t stop mattering, but being the person who knew things others didn’t stopped being such an enduringly valuable position.

Not cleanly, and not all at once, but what rose up was a generation of pattern-finders: quants, framework-builders, systems thinkers, algorithm architects. The people who could look at complexity and derive structure from it, who could see the underlying logic of a market, a workflow, a system, and make it legible and actionable. For roughly the past forty years, maybe a little longer, that capacity has commanded an extraordinary organizational and economic premium. And yes - I know I’m compressing a lot of nuanced professional history here.

Naturally, the people who built careers on knowing things didn’t go quietly. Most argued, reasonably enough, that what they did – what they knew – contained irreducible human depth no framework or machine could replicate, or that they would remain a necessary complement to the pattern-finders. The market seemed to have been less sentimental about that distinction than they were. Information possession is rarely enough to confer authority anymore.

I think pattern-finders may be next.

The Recycling of Patterns

Here’s the more uncomfortable version of what I’ve been turning over in my head: most of what I thought of as pattern-finding as a skill wasn’t exactly that. It was something closer to pattern-collection, then pattern-recycling. The genuine pattern-finders — the people who can consistently look at situations and produce something structurally new — have always been rare. What I think most pattern-finding expertise actually consisted of was people building well-curated libraries and practiced judgment about which item in their catalog fit the current situation well enough.

This is hard to say, partly because I’ve spent my career in rooms full of people I respect who were doing exactly this, and partly because I was, and am, doing it too. The thing is, it’s genuinely hard to distinguish between someone who is retrieving a sophisticated pattern and someone who is creating one. Both can look like originality in the moment. Both may even feel like originality to the person doing it.

What I see AI doing, with increasing comprehensiveness, is showing how much of this work is a kind of sophisticated recycling. Not because LLMs are smarter than the best human pattern-identifiers, but because it turns out that what most professional pattern-finding actually is maps disturbingly well to what large language models do natively: aggregate, interpolate, recombine, and generate fit judgments at scale, cheaply and continuously. Now, I don’t know how much “most” is, and that may be the question AI is actually forcing into view. But it’s hard to argue that a catalog of patterns and a working sense of which one plausibly fits can be pulled, unnervingly well, from the statistical relationships encoded in these systems.

And that somehow feels violating to the kind of expertise that has been treated as critical in organizations for my entire career. Not just because a machine can do it, but because watching a machine do it reveals something about what the expertise was all along.

Organizations weren’t irrational to value this. They rewarded (and for the moment continue to reward) something genuinely valuable and rare for the problems organizations face. Companies were dealing with increasingly intricate challenges: global supply chains, matrixed organizations, fragmenting markets, accelerating product cycles. The people who could recognize “we’ve seen this before; it’s really X” became extraordinarily valuable. Management consulting exploded. Frameworks proliferated. The consultant who could walk into a struggling retail chain and say, in effect, “this is how you’re in a classic innovator’s dilemma problem,” or the designer who could recognize “what you need here is this jobs-to-be-done lens,” commanded a premium because they could make complexity tractable by fitting it into known patterns.

I suspect the people elevated by those mechanisms (including many of the people currently in positions of authority to respond to this moment) have developed deep investments in a world that is now shifting underneath them. I don’t think of this as cynical self-protection, at least not mostly. I think it is more structural blindness. Or maybe ordinary institutional caution, since I am certain that people who have lived through enough supposed revolutions acquire good heuristics not to respond to every one. But here’s the pattern I see: a perceptual apparatus formed by decades of experience in one epoch becomes, at the moment of epoch change, the very thing that makes it harder to see clearly. There’s a tragedy there the Greeks would have relished.

And I include myself in this. I’m literally using pattern-finding to analyze the disruption of pattern-finding. That recursion does not fully resolve.

What Comes After Seeing?

There’s something I’ve been circling around in conversations with people who seem to be less disoriented by this moment. I don’t have a clean label for it yet, which I think is itself meaningful. It’s not knowledge-holding. It’s not exactly pattern-finding either. It may be something more like a formed sensitivity to where current explanations stop working in ways that are interesting rather than merely wrong.

Let me try to be a little more precise, while admitting that this is the part I understand least clearly.

Every applied pattern has edges: places where the algorithm almost works but doesn’t quite, where the framework generates confident but subtly wrong readings, where the map and the territory don’t fully align. Most people, most of the time, smooth over those edges. The pattern is good enough. Decide and move on.

Sometimes people are arrested by these edges. They find the breakage more interesting than the fit. They drag their frameworks into unfamiliar territory not so much to apply them, as to find where they fail. And when they fail in ways that feel charged with interesting follow-up questions, something fires. A sense that there is something here that the explanation doesn’t account for. That the current framing is insufficient, and that a deeper pattern might be available if someone goes looking.

This isn’t intuition in the mystical sense. It’s perception bent, over time, by deep and consequential experience: by having been wrong in ways that required genuine reconstruction rather than incremental repair, by having cared about outcomes in ways that left … residue? I’ve been thinking of it as something like what people mean when they say an experienced clinician knows something is wrong before the test comes back, or when a developer catches a code smell before tracing the dependencies. The perception fires before deliberate cognition does, but it is not pre-theoretical. It is saturated with theory, memory, consequence, and scar tissue.

I suspect this is one of the places where AI’s limitation may actually matter … at least for now. This kind of perception seems to require having been somewhere, wanted something, been wrong about something, and cared about the difference. It has been interesting to think about how that sort of pre-cognitive perception relates to LLM capabilities. LLMs can respond impressively within frames. What I have not seen them do is originate that felt compulsion that a way of looking at a problem is failing in an interesting way. Maybe that, too, will prove more tractable than I think. I can’t rule that out. But for now it still seems meaningfully different from what these systems are best at.

Because what this faculty can do, if it is a real and developable one, is to tell when and how to invoke the machine. Once you can sense where understanding breaks down productively, you are no longer trying to out-pattern the machine. You are doing something the machine still appears to need but cannot reliably generate for itself: identifying which breakages are worth investigating.

At the Edges of the Pattern

There’s a phrase that crystallized this whole rabbit hole for me, and I haven’t been able to shake it. The first professional cognitive identity was: I am someone who knows things others don’t. That was displaced by: I am someone who sees what others can’t. What I think may be displacing that now … what the people navigating this moment most usefully seem to embody … is something I’m still looking for a label to describe.

The phrase: I am someone who senses where understanding breaks down.

Admittedly, a convenient progression.

The Harder Part

I do have to admit this argument has a self-serving structure I can’t fully escape. Every displaced cognitive class has argued they possessed some irreducible human depth that couldn’t be systematized or transferred. The fact-holders said it. The pattern-finders say it. The pattern repeating doesn’t validate the claim, but it definitely makes it seductive.

If I’m even directionally right, the implications are unresolved and probably not comfortable. If the genuinely valuable human contribution increasingly resides in those rarer moments of noticing where something doesn’t fit, the organizational and compensation consequences are profound. A sparse role can be high-leverage without being well-rewarded. Institutions are often better at pricing legible, repeatable contribution than intermittent but catalytic ones. The person who occasionally notices the anomaly that matters may create enormous leverage and still look, on paper, less productive than the person who can generate competent structure and momentum all day long. If that is true, then many organizations may be structurally inclined to undervalue precisely the form of judgment they claim to be prioritizing in this rush to AI.

That, in turn, makes the previous section harder to romanticize. Even if this emergent faculty is real, it may not map cleanly to prestige, compensation, or stable professional identity. It may be rarer. It may be harder to credential. It may be more intermittent than institutions know how to organize around. It may also be a miserable basis on which to try to build a knowledge career. I don’t know.

And the people most unsettled by this shift have every reason to seize on a description of what remains valuable and recognize themselves in it. I’ve tried to write against that pull, but I can’t be sure I’ve succeeded.

What I’m more confident about is this: I don’t think the discomfort in rooms full of smart, experienced people is primarily economic. It is the sense that the thing that they … we … (I?) … have understood to make us value is being hollowed out. The cognitive identity that organized entire careers — I am someone who sees what others can’t — feels like it’s being destabilized at exactly the moment when it is most dominant.

John Henry won the battle. The tragedy wasn’t just losing his life. It was that the victory still proved the role could be defined well enough to be replaced. Everyone watching him fight the machine understood that immediately, even if they couldn’t fully articulate it.

Most of the people going the way of John Henry right now think they are John Henry. What they may be missing is that he at least knew exactly what the machine was doing.

The people who find their way through I suspect are not the ones who become the best prompters, or who bolt AI onto their existing toolkit and call it transformation. They are more likely to be the ones who can tolerate some reconstruction of professional self-conception rather than simply attaching new tools to the old one.