technology

How Much Do I Love You? How Hard Must We Try?

By in ai, bring me the head of nicholas carr, foolosophy, technology, writing No Comments

Derek Thompson:

In my corner of the world, which is journalism and academia and writing in general, Pangram has become nothing less than a cultural force. A meme, even. I’ve seen prominent politicians, chief executives, and public intellectuals exposed as slop-slingers. Their public humiliation sometimes fills me with righteous indignation. But it also makes my toes curl a bit, as I think about people spending all day policing the internet for slop and attempting to Pangram-shame people with a technology that can make mistakes.

Today’s interview is with Pangram founder and chief executive Max Spero. Among other things, we talk about:

  • what if AI writing takes over the internet and nobody cares?
  • how Pangram works—and how and when it makes mistakes
  • why AI writes “like that”
  • my fear that AI-writing witch hunts would be an unwelcome addition to online life
  • Pangram’s relationship with Substack
  • what if a big tech company, like OpenAI or Google, creates their own AI detector that puts Pangram out of business?

Evidently, I’m destined to be a gadfly, though with any luck not the kind that earns a glass of hemlock. I just think that most of what our commentating classes produce is empty verbiage that avoids the ostensible subject. What I find interesting here, as always, is what our choices reveal about our deeper sentiments. Fifteen years ago, when Nicholas Carr was beating the drum over how hyperlinks, search engines and the internet itself were making us dumber, I was zeroing in on an assumption he took for granted. He assumed that we truly wanted to do more difficult tasks like read challenging books, and that we were being prevented, more or less against our will, by technology. I simply noted that this was a big ol’ fat-ass begged question. Maybe all technology did was reveal our base inclinations once given multiple choices.

There was a whole hierarchy of “wanting” worth reflecting upon in Carr’s case. I want to read difficult books. I want to have read difficult books. I want to be capable of reading difficult books. I want to think of myself as somebody who reads difficult books. I want other people to regard me as somebody who reads difficult books. And, at 8:45 tonight, I want to aimlessly scroll around on the internet instead of reading one of the fifteen difficult books I have piled up on the nightstand. Those desires can all coexist, and they tend to be knotted together like last year’s strings of Christmas lights. Untangling them is hard. It’s much easier to blame an outside agent for “rewiring my brain” and preventing me from being my true self.

Now, here we are, where people are driving themselves crazy wondering who is using AI and to what extent, and they’re in agreement that this is unquestionably a bad thing. And once again, I say: isn’t it interesting that the thing we all profess to want is the thing to which none of us apparently want to devote time and energy? To me, all LLMs are doing is finally providing an equal alternative: you mean I don’t have to waste any time trying to write some LinkedIn bullshit? And it turns out that no one actually wants to devote time to writing LinkedIn bullshit!

But — crucially — we can’t bring ourselves to be honest about this, especially when it turns out that we don’t particularly want to devote time to writing posts and essays, either. We fear what it might say about us if we spend most of our lives wishing we could be elsewhere doing other things. We fear an honest accounting of our sunk costs and an honest reckoning with our regretted compromises. We fear an honest look at how much of our lives we would happily automate away if we had the option. Rather than face that squarely and decide what to do about it, we retreat to the earlier stage: proclaim that, say, writing LinkedIn posts by your own unaided effort is a good and worthwhile thing to do, and anyone who doesn’t want to do it is a bad, shameful person. It’s like a religious fanatic who catches himself thinking sinful thoughts and responds with horror and redoubled efforts at prayer and self-flagellation. Pray the AI away is a strategy, I guess, but I don’t think much of its chances.

We have spent ages assigning moral dignity to activities partly because we had to do them. AI is forcing us to discover how many things we called “meaningful work” because, until about five minutes ago, we had no practical way not to do them. As usual, sudden exposure to undiluted self-awareness sends us into a pants-pissing panic. Who are we in the absence of compulsion? How much choice is too much choice? What are my true desires, or are “desires” too ephemeral to bear the weight of a personality anyway? This sort of reflection seems much more penetrating to me than counting em dashes, which is probably why we’re talking about the latter.

Digging For Gold In My Neighborhood

By in technology, the geist of the zeit, writing No Comments

Daniel Oppenheimer:

I’m not naive about why writers write what they write. We do it for the love of the game, certainly, but most of us also need to get paid, whether the essential currency is prestige, attention, career advancement, or … currency. The blogosphere worked as it did, in the ways that I so loved, because it evolved in a way that rewarded writers and aspiring writers for doing that kind of work. When the incentive structure changed, it stopped working.

For the conversation to return, we’ll need a new structure where the right mix of incentives, disincentives, technology, UX decisions, norms, etc. are in place to make it worth our while, as writers, to write to each other in the way bloggers used to write to each other. I don’t expect it to happen, but consider this my plea to the universe (and, what the hell, the Substack Team as well) that it will.

There’s a funny little circularity to a man asking, on Substack, “Can we bring back blogging? I miss it.” Well, you’re not asking this on Blogspot, so I guess the answer is no, at least in the sense you mean it. Which raises the obvious question: why act like a hapless bystander, pleading to an indifferent universe? It looks like you’ve already got danieloppenheimer.blogspot.com set up and ready to go. All you need to do is delete the Notes account, stop relying on Substack recommendations, give up the subscriber count, publish an RSS feed, build a blogroll, and send readers away from your site constantly, accepting that many of them may never come back. Ah…

It’s almost a textbook collective-action problem. Someone wistfully says, “I’d love a decentralized blogosphere.” Then he says, “But I’d rather not give up my 18,000 subscribers.” Then he looks around and asks where the decentralized blogosphere went. All of that illustrates why things are the way they are. The early-oughts blogosphere was something of a gold rush, unsustainable by its nature. The evolving technological development of the time made visible a deposit of talent that the old distribution system had been bad at finding. Under the old arrangement, being an excellent political or cultural essayist wasn’t enough. You also had to get yourself onto the institutional map. That might mean going to the right university, moving toward the Acela corridor, knowing which magazines took pitches, cultivating editors, acquiring credentials, tolerating junior positions, and generally possessing enough career orientation to enter the pipeline. There must always have been people who could write wonderfully but who were badly matched to that process: the insurance adjuster in Nebraska, the teaching professor at a small liberal-arts college in Ohio, the lawyer who had no desire to become a journalist, the housewife with encyclopedic knowledge of some obscure subject, the obsessive amateur reading three books a week. Their talent was economically and culturally stranded.

Then Blogger and Movable Type arrived. And like a gold rush, the first people into the territory could obtain wildly disproportionate returns. If you were unusually good in 2002, there weren’t yet ten million people competing with you for attention. A link from the right person might expose your work to thousands of readers. Those readers might link it again. Within a short period, somebody who had never been through the institutional filtering machinery could acquire a readership, something that previously required an editor to bestow. And a readership was convertible into columns, magazine assignments, book proposals, speaking gigs, and a professional network. Indeed, that’s where most of the talented bloggers ended up. But for every one of them, there are probably a thousand blogs adrift on the Sargasso Sea, abandoned shortly after launch with no signs of struggle.

There’s a sentimental suggestion here that the blogosphere was wonderful because people once possessed better norms of conversation. Well, I was there, and while things were certainly more decentralized, which is generally a good thing, it isn’t hard in retrospect to see how social media was just an intensification of already-existing trends rather than a qualitatively different thing. People linked, posted constantly, picked fights, sought recognition, discovered unknown writers and answered one another partly because there were enormous potential rewards for doing so. The frantic cross-linking Oppenheimer remembers wasn’t merely a lovely communal custom. It was also prospecting behavior. There was gold in them thar hyperlinks. “Can we bring the blogosphere back?” in 2026 sounds slightly like asking “Can we bring back the California Gold Rush?” in 1900. You can move to California and pan for gold. Gold still exists. You may even find some. But you can’t recreate the historical moment when somebody discovered that an enormous previously unknown deposit was sitting there and word had only just begun to spread.

The gold-rush people came to the blogosphere hoping something might happen to them. Blogging was only a means to a professional end. The people still out here twenty years later are the ones who discovered that they rather liked living in an abandoned mining camp after everybody else left. It’s a good life, if you like wide-open spaces, little excitement, and sparse conversation with the cacti, coyotes and tumbleweeds. But if you think a bunch of urban Substackers are going to give up their careers and move back out here, well, I have a deed to a mine I’d like to sell you.

Sculptural Writing

By in ai, aphorisms, books, technology, writing No Comments

Machines are still only in their trial period. One day, perhaps, an electrophonic kaleidoscope will compose musical figures in their thousands — will invent whole series of rhythms and melodies. Man will have machines that create just as he will have machines for precision thinking — His role will be reduced to choosing. Combination factories.

The most intense feeling is only one combination among others.

— Paul Valéry, Cahiers / Notebooks 3

The standard complaint about large language models is that they will make writers sound alike, and perhaps eventually think alike. But Valéry’s prescient suggestion makes me wonder about another possibility. What if these machines do not merely automate an existing kind of writing? What if they also call forth a somewhat different skill in writers?

We ordinarily imagine writing in painterly terms. The writer confronts a blank surface and begins putting things onto it. A word, then a sentence, then a paragraph. The work accumulates. Even when we know perfectly well that revision will follow, our basic picture of authorship remains additive: there was nothing here, and now the writer has made something out of his inner resources and aesthetic vision.

An LLM makes possible a more sculptural way of working. Instead of confronting a blank page, the writer can begin with an excess of material. The machine can produce the boulder of text. The writer’s task is to decide what, if anything, is inside it. That sounds close to Valéry’s prediction that our role will be “reduced to choosing.” But choosing may be doing a lot of work there. A sculptor also “chooses.” This part stays; that part goes. This line needs to emerge; that mass is obscuring it. The finished form depends not on the ability to manufacture marble but on seeing possibilities in it that someone else would not see. Two sculptors can begin with approximately the same block and end with objects bearing no resemblance to one another.

Give two people the same serviceable paragraph from an LLM. One may read it and think it’s fine as is. The other notices immediately that the opening sentence is unnecessary, the metaphor in the middle is the only interesting thing there, the conclusion merely repeats the premise, and one odd phrase suggests a better direction for the whole piece. This would shift some of the emphasis in writing from generation toward discrimination. An LLM can give you ten openings before you have finished thinking about the first. It can propose twenty metaphors, five structures, three counterarguments, a dozen ways of phrasing the sentence you are stuck on. Under those conditions, producing another possibility is nearly worthless, but knowing which possibility is promising becomes more valuable. That sounds less like the traditional ideal of the writer and more like the work of an editor.

I haven’t experimented with this yet. I don’t know how it might go in practice. It probably would not go one single way for everybody. But this seems more interesting than treating AI-assisted writing solely as a question of whether the machine has “done the writing.” That formulation assumes we already know where the important part of writing resides. Well, maybe. Or maybe abundance changes where the difficulty lies. The blank page tests one set of capacities; the overflowing page may test another.

Why Y?

By in ai, foolosophy, technology, the geist of the zeit

Gregory Conti:

AI means the death of authorship—and of the inspiration that work which you know reflects human achievement, and human competence, provides to all who read it. We will increasingly become mere “prompters,” at best “collaborators” with the machine. As sci-fi as it sounds, we are already well on our way to becoming mere interchangeable vessels for ChatGPT’s accomplishments and insights. The awe that we feel in reflecting on an Einstein or a Dostoevsky, we will feel toward no one again; Claude cannot be someone’s hero. The whole concept of a “life’s work” that belongs to a great progenitor of ideas is under threat; soon we may have less reason to care who “wrote” a paper with an AI’s help or who inaugurated a theory by prompting Claude than we do about who was riding Secretariat when he won his triple crown.

Conti is reflecting on what AI means for human achievement in the big scheme of things. I don’t have strong feelings about the majority of his points, but this one stands out like a sore, dumb thumb to me.

The crucial word is interchangeable. Give the same powerful tool to ten different people and they do not become interchangeable unless the only thing they contribute is pressing Enter. They bring different questions, accumulated knowledge, obsessions, standards of evidence, aesthetic preferences, capacities for judgment, willingness to reject attractive nonsense, and, perhaps most importantly, different ideas of what is worth doing in the first place. AI can supply an enormous amount downstream of those choices without making the choices cease to matter.

In fact, Conti’s own essay inadvertently demonstrates this. Claude apparently got close to an argument he was already developing, and he found that “phenomenally deflating.” But why was that particular argument significant to him? Why had he chosen those primary sources? Why was he interested in the question? Why does he accept some conclusions and reject others? Claude did not spontaneously decide one morning that technological succession was a threat to human flourishing and pitch an essay to Compact. Gregory Conti did. The machine’s competence can potentially alter how much credit he deserves for particular inferential moves, but it doesn’t turn him into a fungible nozzle through which Claude squirts ideas.

The machine, unlike Secretariat, doesn’t want to run anywhere. Secretariat was an organism pursuing the same immediate end as Ron Turcotte: get around the track ahead of the other horses. Claude has no intellectual project corresponding to Conti’s. It doesn’t care whether AI destroys authorship, whether the argument is published, whether it is true, whether it contradicts something it said yesterday, or whether anybody reads it. It can perform cognitive labor inside a project, sometimes astonishing amounts of it, but the project still has to come from somewhere.

“Claude cannot be someone’s hero” is therefore true and almost completely beside the point. Neither can a forklift. Yet the existence of forklifts has not destroyed our ability to admire someone who deadlifts 1,000 pounds. A forklift can make the human achievement look absurd if the only thing that matters is “how many pounds moved vertically?” But that was never what the human achievement meant. We admire a human being’s ability and desire to do it.

Cars didn’t destroy running. Calculators didn’t destroy mathematical virtuosity. Cameras didn’t eliminate painting. Recorded music didn’t make live musicians pointless. Mechanical advantage doesn’t automatically annihilate human accomplishment because human accomplishment has never consisted simply in maximizing output irrespective of the means. By Conti’s logic, chess should have become meaningless sometime around Deep Blue, because there was “one fewer arena in which a person, through their diligence and hard work and intellect, can shine.” Instead, millions of amateurs remain devoted to the game, and Magnus Carlsen remains famous precisely because nobody gives a damn that Stockfish could annihilate him.

A person who spends twenty years reading, thinking, developing peculiar preoccupations, deciding what questions matter, interrogating an AI, recognizing which of its suggestions are fertile, rejecting ninety-nine others, connecting one result to something he read fifteen years ago, and shaping all of that into a coherent body of work isn’t an “interchangeable vessel.” He’s doing something much closer to what authors have always done: imposing the form of one mind upon heterogeneous material. Conti sees a real problem — machine contribution can become large enough that conventional human credit becomes fraudulent — and expresses it as the much grander and far less defensible proposition that machine capability renders human individuality irrelevant. Those aren’t remotely the same claim.

AI attacks the scarcity of competence much more directly than it attacks the scarcity of salience. Producing a competent essay, competent illustration, competent computer program, competent critique — these things may become astonishingly cheap. Deciding that this obscure conjunction is interesting, that that apparently trivial experience deserves attention, that everybody else is asking the wrong question, that a ridiculous analogy is actually illuminating: those are different faculties. Weirdness is one of the sources from which importance originates. Dostoevsky is not Dostoevsky because he was extraordinarily proficient at generating grammatical Russian prose. Plenty of people could do that. What mattered was the fantastically peculiar configuration of obsessions that made The Brothers Karamazov seem necessary to him in the first place.

This is also why the “interchangeable vessels” line strikes me as almost exactly backwards. The more abundant generic intellectual competence becomes, the more valuable non-interchangeability may become. If everybody has access to something capable of producing a polished 2,000-word essay on Augustine, then “can produce polished prose about Augustine” stops distinguishing anyone. What remains distinctive is: Why Augustine? Why this passage? Why connect it to something your warehouse supervisor said yesterday? Why does that remind you of an obscure Tolstoy diary entry? Why do you find the conventional interpretation unsatisfactory? Why make a joke about forklifts halfway through?

There was a smart-ass kid in my Philosophy 101 class who argued that science could answer all philosophical questions better than philosophy could. My professor’s quip was that science couldn’t answer the question “Why?” Decades ago, when LLMs were still science fiction, that fault line was already obvious to anyone who looked. Science, technology, AI may quickly and effectively answer the question “Given X, how do I achieve Y?” They may have less to say in response to “Why Y?”

If I had to speculate on future trends, I’d suggest that we are probably headed toward a world in which competence is cheap and personality is expensive. And, amusingly, that may make the intellectual oddball rather more valuable, not less. The person who can write the 400th respectable essay about “AI and the Future of Humanity” is in trouble. The lunatic who suddenly realizes that prednisone, French symbolist decadence, and the ability to easily put on one’s socks belong in the same sonnet still has a job.

Most of these big-picture overviews of AI deal in ventose generalizations rather than the interesting particulars. This makes the genre frustrating and nearly useless to me. A sentence like Conti’s “we are already well on our way to becoming mere interchangeable vessels for ChatGPT’s accomplishments and insights” is nearly evidence-proof. What observation would establish that people had become “interchangeable vessels”? What observation would disprove it? The phrase creates an atmosphere rather than identifying a mechanism.

And because so much AI criticism operates at that altitude, it has developed a remarkably standardized inventory of anxieties: loss of authorship, loss of authenticity, deskilling, homogenization, dependence, replacement, alienation, loss of meaning, death of creativity. The nouns change places, but you can often predict the next paragraph. There is something genuinely funny about essays warning that AI will produce generic thinking while themselves reproducing a highly recognizable genre template.

From GeoCities to the Server Farms, This Land Was Made for You and Me

By in jests japes jokes jollies, technology

I had come to look at French Revolutionary pamphlets when I encountered the fence.

It stood across the whole screen: a black barricade, high and wide, admitting passage through a single white gate marked “Continue.” Before proceeding, I was required to hear an acknowledgment of the traditional owners of “this land.”

This land?

I looked around. I was sitting in Virginia. The National Library was in Canberra. The pamphlets came from Paris. The server was probably humming in an industrial park known only to three technicians and a tax assessor. Yet somehow we had all gathered upon the same sacred patch of ground.

Actually, you know what? You’re on my turf. I am an indigenous person — of cyberspace. At last, a homeland with a reasonably well-documented date of first settlement. No disputed oral traditions, no archaeological layers, no inconvenient migrations extending back beyond the arbitrary cutoff date. Before the white nerd arrived, the blinking cursor grazed upon an empty screen.

Rocks, trees, rivers, mountains — granted, prior claims abound for those. But the JPEG of the rock, the polygonal tree, the river rendered in sixteen colors, the mountain represented by ^ characters? Those arose within our ancestral domain. The native fauna were trolls, worms, daemons, and badly drawn ASCII cats. The sacred sites were Usenet groups, dial-up bulletin boards, GeoCities pages, and forums whose moderators had not seen daylight since 1987.

Indeed, “digital native” may be the one indigenous identity modern institutions recognize without embarrassment, though even here the original inhabitants are gradually being displaced by vast migratory hordes carrying smartphones and asking why every website cannot simply be an app. My people were here before the smartphone settlers came. Before the apps. Before the platforms. Before a billion colonists boarded their little glass vessels, sailed into our waters, and began demanding that every form of human expression be rearranged into a vertical rectangle.

We crossed the wilderness with nothing but AltaVista, typing URLs letter by letter. We had no predictive text to guide us. One misplaced character could send a traveler into the void. Later, some unknown benefactor taught us to copy and paste, an invention still celebrated in our oldest songs. We built our settlements with raw HTML. Every scrap was used. We painted our words red, blue, yellow and green. We tiled the background with tiny stars. We consulted “For Dummies” scrolls to construct tables inside tables because CSS had not yet been bestowed upon mankind. Our shamans summoned special characters through the incantation of hex codes. Our homepages contained photographs of iguanas, quotations from Nietzsche, guitar tablature, collections of animated skulls and exhaustive accounts of why Rush had been misunderstood by the critics. They were not polished, but they belonged to us.

Every settlement had a guestbook. Every village had a message board. Webrings joined distant tribes. A man could depart from a page devoted to medieval siege engines and, after six or seven migrations, find himself among collectors of Soviet toothpaste labels. We called this “surfing,” though few of us had ever been near an ocean. Our sacred structures were marked UNDER CONSTRUCTION, which expressed our cyclical understanding of time. Nothing was ever finished. Nothing needed to be. A page might remain under construction from 1998 until the death of the sun.

Then the settlers came.

At first they arrived in manageable numbers. They built blogs and learned our customs. Some even adopted native names such as XxShadowWolf77xX. But the great migration began with the smartphone. Suddenly millions poured ashore without keyboards, mice or any knowledge of directory structures. They did not explore the web. They expected it to cater to them.

The open country was enclosed. Forums were abandoned and replaced by imperial platforms. Personal homepages gave way to standardized profiles. The old paths disappeared beneath algorithmic roads that carried every traveler past the same attractions. Then they began appropriating our culture.

They discovered dark mode, as though our ancestors had not read green letters on black screens since time immemorial. They invented “online communities” after destroying thousands of forums and replacing them with Discord servers whose contents vanish from public memory. They praised “digital minimalism” while carrying devices that screamed for attention thirty times an hour. They turned the personal homepage into “personal branding,” stripped away the iguana, the MIDI file and the purple text, and sold the remainder as professional identity.

And now, after all that, I find a fence across my ancestral land. Where is Midnight Oil with a ballad dedicated to my people?

Happy Willis Carrier Day!

By in environment, technology

Ed West:

Until relatively recently the world was dominated by a handful of relatively cold regions, and Paul Johnson observed in his history of the United States that human industry thrives in what Fahrenheit appreciators would call ‘the 60s’, a Goldilocks zone that turned New England and Greater Yankeedom into a powerhouse. The southern states, in contrast, were held back by higher mortality and the impossibility of productive work in the sweltering summer months, and until the mid-20th century were about 40 per cent poorer than the Union states.

The northerners were especially known for their work ethic and their inventiveness, a characteristic epitomised by the Boston-born Benjamin Franklin. Then in 1902, New York’s Willis Carrier changed everything with the one of the most world-changing inventions, the air conditioning unit; aircon has notably shifted population and power in the US southwards, but Dixieland is just part of a broader, global ‘sunshine belt’.

Fools and Tools

By in ai, foolosophy, technology

L.M. Sacasas:

But AI is not a tool in this sense, it is an environment which envelops the user and works on us from the inside out while we naively think that we remain unchanged by our use so long as we are using it carefully and intentionally. The care and intentionality is beside the point, and our confidence in such vigilance probably works against us in the long run.

Sacasas’s warning is aimed precisely at people like me: the careful user, the intentional user, the person who thinks he has bounded the machine and knows what he is doing with it. Very well, he has my attention. But a charge this serious requires more than prophetic posture. If my vigilance is part of my delusion, show me the delusion. If my careful use is deforming me, show me the deformation. Tell me what faculty is being weakened, by what habit, through what mechanism, and by what signs I might recognize the damage. Instead, the essay largely lets pugnacity and citation do the work of argument. McLuhan and Illich are summoned like heavy artillery, but the target remains vague. The reader is meant to feel that something deep has been proved because the vocabulary is grave enough. But the key claim, that careful, intentional use is not merely insufficient but self-deceiving, is asserted rather than demonstrated.

The eternal missing question: compared to what? He writes as if the alternative to AI were some richer, slower, more embodied intellectual life waiting patiently nearby. In my case, the alternative is often that the conversation does not happen at all. A stray intuition appears during a hectic week, flickers for a moment, and then disappears under work, fatigue, errands, training, and the ordinary attrition of life. AI does not replace the seminar, the monastery, the long walk with a learned friend, or the silence of the study. Those were not on offer. It supplies kindling for sparks that would otherwise go out.

The question, then, is not whether AI forms us. Of course it does. So do books, notebooks, friendships, instruments, gyms, marriages, and tools of every serious kind. A barbell is not “neutral” in the sense of being consequence-free; used stupidly, it can deform the body. But no sane person concludes from this that barbells are therefore not tools, or that careful use is only fooling oneself. The conclusion is that dangerous tools require form, limits, instruction, humility, and disciplined attention. Sacasas wants to deny the slogan that technology is “just a tool,” but he overcorrects into a different simplification: that because AI is formative, agency is naïve. The most charitable framing might be that the vast majority of people don’t use AI the way I use it. Fair enough if true, but I don’t know how we could quantify a “median use.”

The Suspicion Is Coming From Inside the House

By in ai, jests japes jokes jollies, technology, the geist of the zeit

Blurbs? Formulaic and vapid? The hell you say!

The suspicion is not coming from the machine. It is coming from the people who have chosen to make AI into a moral pollutant. Once you decide that certain textures of prose are spiritually contaminated, you shouldn’t be surprised when you begin seeing contamination everywhere, including in your own work. AI has not made you paranoid; your vendetta against AI has. You have trained yourself to read for impurity, and now you cannot stop.

Old Novelties

By in extraordinary popular delusions, media/propaganda, technology

Nick Kristof:

You want security, health care and the American dream? Look to Scandinavia.

“We actually live the American dream,” Jens Stoltenberg, a former prime minister of Norway who is now the finance minister, told me. “The American dream, it’s more reality in the Nordic countries than in America.”

On the Fourth of July, The New York Times invited Americans to contemplate the superior realization of their own national dream in Norway. There was something almost comfortingly traditional about it. Every American holiday requires at least one progressive reminder that another country performs America better than America does, just as every Thanksgiving requires an essay about genocide and every Christmas an argument over the political implications of the Nativity.

The Scandinavian comparison is now older than many of the people presenting it as an urgent new idea. Why can’t America have Norwegian wages, Norwegian vacations, Norwegian child care, Norwegian health care and Norwegian social trust? We simply take the good features of Norway and install them here, as though nations were operating systems and social arrangements downloadable patches. Ten years on from the last time I bothered thinking about any of this in print, I see progressives still haven’t accepted any updates to their worldview.

The objections are predictable because the fantasy is predictable. Norway is small, wealthy and historically homogeneous. Its welfare state grew within a society whose members possessed a strong prior sense of belonging to one another. Its citizens accept broad taxation, powerful labor institutions, extensive public administration and social expectations that Americans might find stifling. Its present arrangements are the product of a particular history, population, culture and moral compact. They cannot be detached from that surrounding life and shipped abroad like barrels of North Sea oil.

But the more interesting question is why this familiar vision retains the aura of innovation. Human beings are psychologically conservative creatures. We enjoy imagining the future, provided it resembles the present closely enough to be reassuring. We permit ourselves to remove everything we dislike — poverty, inequality, insecurity, bigotry, tedious work — while preserving everything familiar and desirable. The future becomes the existing world with its blemishes digitally erased. What we find harder to tolerate is genuine novelty: developments that alter the furniture rather than rearranging it.

Artificial intelligence has exposed this conservatism with unusual clarity. The same cultural class that spent decades celebrating disruption, innovation and the coming knowledge economy now publishes an apparently endless succession of articles lamenting what AI may do to education, art, employment, authorship, journalism, friendship, democracy and the human soul. Some fears are sensible; others are merely atmospheric. Taken together, they reveal how little desire most people have for a future that has not already been morally domesticated.

AI is unsettling because it does not offer a purified continuation of the familiar order. It threatens to change who possesses knowledge, which skills command status, what counts as authorship and whether many respectable forms of symbolic labor remain scarce enough to support professional identities. It may not merely improve the existing world. It may make parts of that world obsolete.

Scandinavian social democracy poses no comparable imaginative danger. It is an old novelty — an inherited alternative that still allows its admirers to feel adventurous. Its lineage is familiar, its moral language is settled, its villains are recognizable. Raise taxes, strengthen unions, expand benefits, regulate markets, empower competent administrators. These ideas may be good, bad or mixed, but they are not new. They belong to a political imagination formed generations ago. Their age is part of their appeal. They offer the thrill of futurism without the terror of the future.

“Norway” signifies the same consoling possibility: modernity without its drawbacks, capitalism without insecurity, diversity without conflict, government without incompetence, solidarity without conformity and prosperity without tragic choice. This is less a policy program than a spiritual aspiration. It expresses a wish to live in a world where every good thing has been retained and every bad thing removed. When someone asks what must be sacrificed, which disciplines must be accepted or which goods may prove incompatible, the question sounds almost indecent. Why introduce trade-offs when compassion has already shown us the destination?

These arguments so often begin with moral spectacle rather than institutional explanation. Look at the waiter earning twenty-five dollars an hour. Look at the cheap day care. Look at the five weeks of vacation. Look how happy everyone is. The reader is invited to admire the results before being asked to consider the society that produces them. Everyone loves the sight of Japanese sports fans cleaning up the stadium after the game, but how many emotionally-incontinent progressives would accept the childhood training, strong expectations of conformity, culture of shame, deference to group norms and comparatively little tolerance for the person who says, “I don’t feel like it” necessary to produce it? Likewise, Americans who celebrate Nordic generosity rarely dwell on the middle-class taxes, social conformity or administrative reach because those are not the spiritually gratifying parts.

The habitual fantasy is to select national virtues à la carte: Japanese public conduct with American individualism. Nordic security with American taxation. American dynamism without disruption or inequality. Mass immigration without assimilation. Radical diversity with high interpersonal trust. Universal solidarity without a strong national identity. When one of these combinations fails, the failure is treated as evidence of insufficient compassion, bad messaging or sabotage by reactionaries, not as evidence that the desired goods may inherently conflict with one another. The aspirational version suppresses those conflicts because it wants moral credit without tragic choice. It imagines politics as the accumulation of goods rather than the ranking of them. But governing is mostly deciding which imperfection you can live with.

Are You Finished?

By in ai, technology, writing

Adam Kirsch:

And that is why the rise of AI writing represents a great opportunity for literature, even as it makes life harder for professional writers. When photography was developed in the 19th century, it replaced painting for most utilitarian purposes; a camera could document what things looked like more accurately and cheaply than a painter could. But the art of painting didn’t die out. On the contrary, it entered a golden age: Freed from the obligation of realism, painters developed radical new ways of seeing, such as Impressionism, Cubism, and abstract expressionism. Now AI has the potential to liberate literature in the same way. In a world full of emptily competent prose, we need writers daring, challenging, and obstinate enough to tell us what it’s like to be human, “from the inside.”

I have to admit, I’ve been losing a lot of respect for writers as a species lately. Apparently a sedentary life of scribbling has turned their vertebrae to jelly and left them cowering in the face of a challenge. Most of the responses I’ve seen to the prevalence of LLMs have reminded me of Private Hudson in Aliens:

It’s a relief to see Kirsch, whom I’ve always appreciated, being the voice of common sense here. Having a strong point of view, the one thing an LLM can’t have, goes a long way toward preventing one’s writing from having a beige-tapioca sameness to it, even if AI is involved in the making of it. Kirsch amusingly notes that many of the bot-sniffers sound like phrenologists in their insistence on diagnosing LLM degeneracy by the telltale skull shape rhetorical devices. It is true that LLMs give fluent shape to half-thought. A weak writer with no depth or perspective will produce clichés artfully expressed, which is what most people are reacting against: all flash, no substance. But the rhetorical devices can survive overuse when the thought has blood and sinew in it. People don’t get tired of parallelism, antithesis, repetition, climax, reversal, aphorism, or rhythm as such. They get tired of being nudged toward the feeling of significance where no significance has been demonstrated.

A scale is not a song, but without scales, or something scale-like, most music collapses into gesture, noise, or private experiment. Likewise, chiasmus, tricolon, epimone, polysyndeton: these are not cheap tricks in themselves. They are ways of organizing attention. They let the reader feel relation, contrast, emphasis, hierarchy, arrival.

The current suspicion toward certain rhetorical devices, another Hudson-like hysterical overreaction, is mostly a confusion of cause and symptom. People are seeing a lot of AI prose with em dashes, antitheses, and neat moral reversals, so they start blaming those devices. But that is like blaming I–V–vi–IV chord progressions for bad pop songs. The progression is not the enemy. The enemy is dead usage, automatic usage, usage without pressure from perception. The existence of Taylor Swift’s career doesn’t discredit the tonal system and force us to pretend to enjoy Schoenberg. Abandoning rhetorical structure because AI overuses it would be like abandoning melody because jingles exist.

Most prose is not literature. Most prose is paperwork, filler, correspondence, summary, instruction, polite institutional lorem ipsum. AI will colonize that stuff easily because much of it was already written as if nobody in particular had written it. But literature, essays, criticism, aphorism, memoir — anything where the value lies in the irreducible angle of attention — may become more sharply itself. The danger, maybe, is that many readers will be perfectly content with dispenser-literature, just as many people are content with chain restaurants, algorithmic playlists, prefab opinions, and cinematic universes. But that is not new either. Literature has always had to coexist with entertainment, sermon, formula, and narcotic. AI simply removes the last excuses. If the machine can write the thing you were going to write, perhaps the thing was not yet yours.

Our Remedies Oft In Ourselves Do Lie

By in ai, bring me the head of nicholas carr, technology, writing

Andrew Potter:

But the hard work of thinking is literally what makes us human. System 1 is your primate brain, the legacy of our evolutionary heritage from the ancient savannah. The hard, focused attention that characterizes System 2 thinking is a currency, and what you choose to spend it on defines what you care about, and ultimately, who you are. Which is why whatever else AI does, whatever benefits it brings the world, there is a sense in which it literally steals your soul.

To me, this is an updated version of the Nicholas Carr-type argument that the internet/social media/AI is “rewiring our brains.” From my perspective, if your attention span and “love of reading” didn’t survive the encounter with Angry Birds, it wasn’t very strong to begin with. If your love of “writing and thinking” doesn’t survive the encounter with AI, perhaps you have some painful introspection to do. If anything, I love writing and thinking more now because I have an interlocutor who never gets tired or bored and helps me to build out thoughts that would have remained unfinished otherwise. AI feels to me like I’ve been handed a Zippo lighter. Formerly, I had to bang rocks together, or rub sticks together, hoping for a spark of insight to catch. Now, the fire is blazing much quicker, but the heat is the same as the “organic” version. The “artificial” heat still allows me to do all the things one normally does when one’s teeth aren’t chattering and one’s fingers aren’t numb. AI has not “replaced” the fire. It has replaced some of the fumbling around in the cold, the freezing in the dark while proving your authenticity to the indifferent woods.

In some sense, every moment is a stage for the same eternally-recurring choice: do I make the effort to do this difficult thing, or do I make an excuse to not do it? It could be reading abstruse books, it could be lifting weights, it could be anything that forces the challenge upon us: this will take time and energy; do I want to do it? This choice is never resolved for once and for all. There is no “progress” here. If I give in to laziness and stop lifting weights, I don’t get credit for all the time I put in; I only succumb to entropy. Likewise, if I let an LLM do my writing and thinking for me, my mind turns feeble. If I recognize it happening and fail to change my habits, well, then I fail, period.

I hear complaints like Potter’s and I just shrug. This is the human condition; what do you want me to tell you? You want me to remove all temptation for you? You can’t handle the truth of the choices you make, so you want all the nearby mirrors covered so you don’t have to look yourself in the eye? Yes, it’s always pleasant to have a burden lifted off of you, at least for a moment. But you simply must philosophically commit yourself to some kind of burden. It’s up to you to choose which one.

Technology is very good at neatly separating your aspirations from your actual priorities, however embarrassing that may be for our self-image. The person who “loved reading” until the internet came along may have loved the identity of being a reader more than the activity. The person who “loved writing” until AI could do it faster may have loved the status of having written more than the labor of trying to say something. If the self disappears the moment an excuse becomes available, perhaps the self was already negotiating its surrender.

Premodern life made many choices for you. You rose when there was work to do because hunger, weather, animals, crops, family, illness, and scarcity were not interested in your self-actualization. Burdens were not “meaningful” because people had nobly chosen them; they were meaningful only in the blunt sense that failing to bear them had consequences. Civilization softens that coercion. It gives us heating, refrigeration, medicine, literacy, machines, contraception, abundance, leisure, mobility, digital tools. Good. Thank the gods for it. But then the question becomes more exposed: now that you don’t have to live under necessity, what necessity will you impose on yourself?

Many people start looking wistfully back toward constraint. Not because they truly want cholera, crop failure, childbirth mortality, untreated tooth abscesses, and hauling water. They want the moral simplification of necessity without the misery that came with it. AI intensifies this because it reaches into the last prestige zone of voluntary effort: thought. We already made physical survival easier. We made social communication easier. We made entertainment effortless. Now even the internal friction of composing, reasoning, summarizing, and articulating can be smoothed. That doesn’t mean the soul has been stolen. It means the soul is being asked, rather rudely: “Were you serious?” The old world forced people to endure hardship. The modern world forces people to decide whether they believe in anything strongly enough to endure hardship voluntarily.

There’s a respectable version of “I need external limits,” and then there’s the evasive version, where a person has enough clarity to describe the trap in exquisite detail but somehow not enough clarity to move the cheese. “I notice that this thing makes me miserable, consumes my attention, weakens my patience, and leaves me ashamed afterward.” Good. But the next step does not require a graduate seminar in attention ethics. Delete the app. Put the phone in another room. Use the browser only on desktop. Set a timer. Make a rule. Break the rule, notice it, revise the rule. Go outside. Read ten pages before opening the laptop. Write badly for twenty minutes before consulting the LLM oracle. None of this is mysterious. It is boring, which is precisely why people would rather inflate it into metaphysics.

This is partly why I talk about the gym so much. Lifting weights is applied philosophy for me. It is a controlled environment where you can practice, in a small way, making the eternally-recurring choice to work harder when you’re tempted to rationalize excuses for weakness and laziness. The barbell sits there, mute, impassively ignoring your pleading and bargaining. You either lift it or you don’t. No serious gym-goer would earnestly claim, the way clever bookish types do, “The existence of couches has stolen my quadriceps.” Couches are always there. Laziness is always there. The body is always trying to economize. The question is whether you willingly resist. If my triceps are weak and making it hard to bench press, what’s the answer? Work on the triceps in particular with supplementary exercises. Not, “society should redesign the gym to suit the needs of people suffering from Tricep-Deficiency Disorder, and we need an awareness-raising campaign…”

The same goes for the mind. The blank page, the abstruse book, the hard sentence, the boring revision, the honest objection — these are forms of resistance. AI becomes dangerous when it is used to remove resistance entirely. But it can also be used like a spotter, a sparring partner, a stack of plates, a second mind pressing back.

I don’t deny the structural question. Someone has to think about defaults, incentives, laws, schools, platforms, and public health. But that is not my interest. I am not a would-be social planner, nudger, or shepherd of populations. I say, drolly, that I’m more of a personal trainer for the soul. I am interested in the person who still has enough agency to be addressed as a responsible individual. I do not want to arrange his cage more humanely. I want him to notice the open door. I want him to become self-sufficient, to take an active hand in shaping his own life, to acquire a taste for chosen difficulty rather than cheap release. The question that interests me is not how society can be engineered around weakness, but how weakness, once recognized, can be trained. The social planner asks: how do we organize society so that weak people do less damage to themselves? The trainer asks: how do we make weakness ashamed of staying weak?

The structural frame becomes dangerous when it gives people a respectable language for waiting. “The incentives are bad.” True. “The platforms are engineered for addiction.” True. “The social environment is hostile to attention.” True. But then what? Are you released from action until the Ministry of Humane Defaults issues its findings? “What do you want me to do? They made this stuff addictive. Until society fixes the problem, I have no choice but to continue doing the thing I have just identified as ruining me.” This person knows enough to diagnose the machinery, but somehow not enough to put the phone in the drawer. He can explain dopamine loops, variable rewards, dark patterns, algorithmic capture, and late-capitalist attention extraction, but cannot survive ten minutes without refreshing the slot machine.

We are never free in some abstract, frictionless sense. We are born into bodies, temperaments, habits, economies, families, technologies, desires, and eras we did not choose. But inside those constraints, there is almost always some movable piece. A small refusal. A rule. A delay. A substitution. A chosen inconvenience. A door left unopened. A tool used only after the first draft. Ten pages before the screen. The phone outside the bedroom. Dumb little measures, maybe, but dumb little measures are what self-command is made of. The structural analysis may be true and still be useless to the person using it as a sedative. Yes, the platform is engineered to keep you there. Yes, the incentives are bad. Yes, your appetites have been studied, modeled, and monetized. But unless you are literally chained to the machine, the next question remains yours: what are you going to do with the portion of freedom left to you?

I Fake It So Real I Am Beyond Fake

By in ai, foolosophy, technology, writing

Krzysztof Pelc:

My wager that we will turn to the past, that we will fetishize pre-GPT work, that we will manufacture new old things, may be wrong. Instead, we may simply adapt. We may come to read the way we now eat—content to consume highly processed fare, vaguely aware of what has been lost, but willing to trade it in for abundance and ease. The label says “homemade flavor,” and that might be enough. After all, while mass production gave rise to a cult of the handmade, industry still won the long game, by imitating just enough of its trappings. Jeans arrive pre-torn, boots pre-scuffed, tables pretreated with the patina of imagined family dinners. If that is the pattern, then the future of literature may not be defined by anxious humanism at all. Perhaps a newly devised look of authenticity will suffice.

Those fake ruins scattered through 18th-century English gardens weren’t really meant to fool anyone. Visitors knew they were built yesterday—it didn’t matter. The feel of antiquity was enough. Same was even true of Ossian. Doubts about the poems’ authenticity surfaced almost immediately; Samuel Johnson called them a fabrication. They swept Europe anyway. Napoleon still declared Ossian greater than Homer. Perhaps we will shrug and just learn to enjoy the fake ruins.

Pelc’s essay is the best I have read yet on what AI use means for those of us who value writing, partly because he avoids both the techno-utopian shrug and the guild-panic howl. He is not asking whether machines will “kill” literature. He is asking what happens when writing survives but provenance becomes unstable.

We often say we want the real thing. What we often want, in practice, is the sensation the real thing gives us. Antiquity without waiting. Labor without calluses. Ruin without catastrophe. Folk memory without surly peasants. Rustic tables without generations of family use. Distressed jeans without years of crawling under cars, stacking lumber, or falling off bicycles. “Homemade flavor” without anyone at home.

This is not mere consumer hypocrisy. It is an old feature of the divided will. “The good that I would, I do not.” We claim to want books, attention, discipline, craft, contact with real minds, slow-grown judgment. Then we spend our hours scrolling, skimming, sampling, grazing. Most of us have more time than we admit for the things we claim to value. We use it badly, and then treat our own habits as if they were acts of fate. The machine did not besiege the fortress of humanity. It strolled through an open door.

Much of the alarm about AI assumes that readers will urgently demand proof of human origin, and some will. There will surely be a market for literary organic produce: free-range essays, grass-fed poems, non-GMO memoirs, certified pre-GPT fiction. Writers will advertise their purity. Publishers will rediscover backlists with priestly zeal. “Lost” manuscripts will turn up in attics, storage bins, dead relatives’ boxes, and obsolete hard drives. A novel supposedly written in 1997 and sealed away for decades will turn out, after enough people have praised its uncanny relevance, to have been fabricated in 2026 with machine assistance. There will be modern Ossians because authenticity, once made precious, becomes profitable to counterfeit.

Some of this will be sincere. There will be readers who want to support human work against machine abundance, just as some people prefer local farms, hand-thrown pottery, letterpress stationery, or actual old furniture. Some will pay a premium because they believe provenance matters. Others will pay because they enjoy being the sort of person who pays for provenance. The distinction is not always easy to draw, even in oneself.

But a larger market may not care. Or rather, it will care only until caring becomes inconvenient. If a bot can more reliably produce competent genre fiction — the cozy mystery, the space opera, the enemies-to-lovers romance, the military thriller, the haunted-house story with respectable pacing and a twist in the third act — many readers will ask the obvious consumer question: did I enjoy it? The plot moved. The characters served. The atmosphere arrived on schedule. The dragon fell, the detective solved, the lovers confessed, the village healed. Another satisfied customer. Why interrogate the supply chain?

Copernicus removed us from the center of the cosmos. Darwin removed us from the privilege of separate creation. Freud, or at least the psychological tradition that followed him, complicated our faith in self-mastery. The sovereign “I” became a disputed territory of drives, memories, habits, compulsions, and unacknowledged motives. AI now presses on what many people had assumed was a safer chamber: expression. Machines could calculate, manufacture, sort, optimize, and predict. But surely they could not make art. Surely they could not speak from inwardness. Surely they could not produce the marks of a life.

Now the answer is less reassuring. The machine does not need to have lived if it can reproduce the marks of having lived well enough to move us.

That is why the bad AI writing we call “slop” is almost comforting. Slop lets us keep the distinction intact. It allows us to say: see, the difference is obvious. The thing has no blood in it. It has that smoothed-out, overlit, hotel-lobby quality, all polish and no pressure. It speaks in prefab sympathies and sterile emphases. It has never noticed anything except the statistical likelihood of what people tend to say after other people have said similar things.

But the unease begins when the writing is not slop. When it has the right hesitation, the right embarrassment, the odd detail that seems rescued from memory; when it does not merely produce cliché but has learned how to imitate resistance to cliché. At that point, the humanist reply cannot simply be, “No, you can’t.” Increasingly, it can.

The problem is that human inwardness only becomes shareable by taking form, and form is exactly what can be copied.

Private experience is not transmissible in itself. I cannot directly hand you my grief, boredom, shame, love, childhood, or sense of a summer evening in 1983. I have to translate it into signs: words, images, rhythm, anecdote, metaphor, gesture, style. If the translation works, the private thing crosses the distance between us. You read the sentence and say, yes, that is how it is. One sealed consciousness has reached another by way of marks on a page.

But the moment inwardness becomes communicable, it becomes patterned. Once patterned, it becomes imitable. Once imitable at scale, it becomes machine-readable. What cannot be expressed remains private, perhaps even meaningless to others. What can be expressed can be copied. This doesn’t make expression worthless but it does make it vulnerable. The bridge between souls can also be used by mimics.

Science proceeds by stripping away the accidental. It wants the reproducible principle beneath the instance. Forget this apple, this eclipse, this falling body: what law governs all such cases? The particular becomes a doorway to the general. But art often moves in the opposite direction. The point is not “a mother,” but my mother. Not “childhood home,” but the house with the cracked step, the basement smell, the June light in the kitchen. Not “romantic attachment” in the abstract, but this person, this voice, this argument in the car, this forgiveness, this silence afterward. The particular is not evidence for the law. The particular is the thing itself.

AI irritates us because it takes the signs of particularity and makes them general. It studies millions of actual human instances — poems, apologies, jokes, eulogies, novels, love letters, confessions, familiar essays — and learns the transferable shape. Then it gives back a new instance bearing the marks of specificity: the cracked step, the remembered kitchen, the awkward pause, the private grief. But those details are not anchored in a life. They are particularity-effects. The formula has learned how to imitate the resistance to formula.

AI drags the old philosophical problem of “other minds” out of the seminar room and drops it into emails, student papers, dating profiles, condolence notes, poems, corporate apologies, memoirs, and love letters. The irreducible thing may still exist. But its outward traces are no longer secure. Perhaps that’s why debates over AI feel so much like debates over free will and determinism. The technical question is never merely technical. We are not only asking whether a system can generate competent prose. We are asking whether the human being can be reduced to math, mechanism, pattern, formula, predictability. Are my choices mine? Is my voice mine? Is my inward life a mystery, or only a process not yet fully diagrammed? The heat of the argument comes from the suspicion that another sacred boundary is being moved.

Many people will still prefer the substitute. If culture is merely the production of desired effects, AI has a strong claim. Let the machines generate our cozy mysteries, breakup songs, fantasy quests, inspirational captions, therapeutic monologues, dinner-party toasts, and grief simulacra. Let each of us float in a private syrup of optimized feeling. If pleasure is the final court of appeal, provenance becomes a luxury concern: nice for those who can afford it, irrelevant to those who only want the thing to work. But if culture is also a relation between persons, then provenance is not a decorative label. It is part of the event. The issue is not whether the fake ruin looks beautiful at sunset. It may. The issue is whether beauty was all we came for.

That question has no easy answer because we ourselves do not live by a single answer. We want the real and the convenient, the handmade and the overnight delivery, the slow book and the fast feed, the soul and the shortcut. AI did not create this division. The deeper fake may not be the forged manuscript or the synthetic essay passed off as human. It may be our own claim that we always wanted authenticity more than effect. Very often, we wanted the feeling of authenticity, and we were willing to take it wherever we could get it.

Acts of Enclosure

By in foolosophy, technology, the geist of the zeit

David Cain:

Just as it’s still possible (though seldom necessary) to ride a horse, it is still possible to surf the internet. It’s a thrill not yet lost to time.

By “surfing the internet” I don’t just mean going online. I mean exploring the internet solely by following hyperlinks from page to page, with no clear destination except for that one wonderful, as-yet-unknown website that will amaze and enthrall you when you find it, the one that will seem like it’s been waiting for you your whole life and which you can’t get enough of.

To surf, you must begin on a normal website with outbound links, and avoid all the algorithm-driven thoroughfares (Reddit, YouTube, X, any “apps”) that direct most of today’s internet traffic. You also have to be on a real computer, not a phone. If you end up on social media, you’re no longer surfing.

Younger readers may not even know that the internet used to be made entirely of websites, created by human beings, connected only by hyperlinks. Hyperlinks served as signposts, hand-placed by other humans, to point fellow travelers to unique locations they would not otherwise have known about. There were no corporate-owned thoroughfares, just many pathways shooting off from each clearing, marked by these handmade signs, beckoning you onward to some other place in the wilderness.

I still browse the web the same way I did twenty-five years ago. I have bookmarks saved in a browser, like a man preserving family recipes in a storm cellar. Even when I use a walled garden like Twitter, I don’t “follow” anyone; I bookmark the “replies” pages of people I find interesting and visit them when I feel like it. I go to sites directly. I remember addresses. I return to places. I did not fully realize this had become antiquarian behavior until recently. I am old. I am a dinosaur. Or, perhaps more accurately, a middle-aged blog badger, nosing around the hedgerows while everyone else is being conveyed somewhere by rail.

Andrew Blackman helpfully summarizes the decline and fall:

  • Social media took off. More people began getting their news and updates on Facebook or Twitter, which were easier to use and deliberately addictive.
  • We started watching more videos and using smartphones and tablets to access the web, and RSS readers tended to be slow to respond to those shifts.
  • The scarcity mentality was now long gone—good information and entertainment was everywhere. We had left bookmarks behind and now lived in the age of Google. Why follow a site when you could just google it if you needed to find it again?
  • Big tech companies realised they couldn’t make money out of RSS and killed it. Google bought the massively popular Feedburner and slowly killed it, before killing its own Google Reader in 2013. Both Facebook and Twitter offered RSS feeds to let you follow people in the early days, but both killed off that feature in… yes, you guessed it, around 2013.

The current internet has plenty of content, but much less place. On a phone especially, the web is no longer a landscape so much as a tray of apps. Facebook is over here, Instagram over there, YouTube in its stall, TikTok in its padded room, Reddit in its sprawling basement, each one sealed in its own atmosphere. You don’t start from a homepage and follow paths outward. You open an attention appliance and wait. The feed replaces the path. The recommendation engine replaces memory. The app icon replaces the address bar. The old impetus was, “Where do you want to go?” The new one says, “Sit still; we’ll bring you something.”

This isn’t a story of prohibition. Nobody forbids you to browse the web the old-fashioned way, just as nobody forbids you to ride a horse to the store. But people will correctly infer that you are making a choice against the age. The old fields still exist. Personal sites still exist. Blogs still exist. Archives, forums, hobby pages, oddball newsletters, marginal cranks, unprofitable enthusiasts are all still out there. The train line now runs through the village, though, and most people do not get off the train and wander toward the fields. Why would they? The lights are on in the carriage. Everyone they know is aboard. The snacks keep arriving.

The platforms didn’t need to destroy the open web. They only needed to capture the habits of movement. The commons were not burned; they were bypassed. The old paths became inconvenient, then eccentric, then faintly invisible. The web beyond the platforms is still technically accessible, but it no longer lies on most people’s daily route. To reach it, you have to interrupt the default motion. You have to get off, climb a stile, and remember that walking is allowed.

I don’t want to make too much sociology out of this, but I suspect some of us carried older habits into the early web because we had been trained in older forms of wandering. Gen X childhood was, for many of us, still free-range enough to make aimlessness feel natural. Latchkey houses, bicycles, vacant lots, woods, arcades, malls, pickup games, aimless wandering. We were sent out into the day and expected to come back at dusk more or less intact. This wasn’t always romantic. We were not all living in a Spielberg movie, backlit by suburban wonder. Often we were just bored. We circled cul-de-sacs on our bikes. We climbed trees. We splashed in puddles. We loitered in woodlots. We played half-organized games that dissolved into arguments. We learned shortcuts, hiding places, drainage ditches, vacant lots, the houses where someone’s mother might give you a popsicle, the houses where a large dog made further inquiry inadvisable.

I remember the drainage ditch in my front yard after a hard rain. I would rush outside with my G.I. Joe figures and stage entire campaigns in the rushing water. Mud banks became cliffs; brown runoff became white-water rapids; a culvert became a portal to mortal peril. It was not a grand adventure. It was just a ditch. But no one had prepared it for me, packaged it, optimized it, or pushed it in front of my face. I had to bring something to it. I had to supply the drama. The ditch didn’t flatter me, guide me, notify me, gamify me, or ask me to smash a like button. It just ran with rainwater and waited.

The early web suited that kind of mind. You started with one site, followed a link, checked a blogroll, landed on a personal page last updated in 1998, found a shrine to some defunct band, followed another link to a crank’s essay, saved a bookmark, forgot what you were originally looking for, and emerged an hour later with three new trails to revisit. It wasn’t efficient, which was part of its value. Foraging is not shopping. You don’t go out knowing exactly what you will bring home. You nose around.

By contrast, the platform web is helicopter parenting for adults. It supervises idle time. It sees the first flicker of boredom and rushes in with a tray: here is a video, here is a recommended post, here is a notification, here is a fight, here is an ad, here is a button to press so you can prove you are still there. It does not trust you to wander, and after a while you may stop trusting yourself to wander.

You can see the training begin early. A child fidgets in public, and a tablet appears. The first signs of unmanaged consciousness are treated like a small fire to be smothered. I understand the appeal. No one wants a shrieking child in a restaurant, least of all the parents sitting closest to him. But there is a difference between occasionally keeping the peace and teaching a child that every blank interval must be plugged immediately with stimulation. Boredom is not a medical emergency. It is the raw material from which attention learns to make its own furniture.

We used to know this. A child with a stick and a creek may not be doing anything impressive, but he is doing something. He is learning how to inhabit time without being pandered to. He is making a world from scraps. He is discovering that the mind can meet emptiness without panic. The platform, by contrast, is preventing attention from growing muscles. It fills every gap before curiosity can form. It replaces “What should I do?” with “What will they give me next?”

Now the content cafeteria has largely replaced foraging, because convenience is powerful, and because convenience becomes culture when enough people accept it at once. The big platforms are easier. They are on the phone. They know who you know. They remember what held you yesterday and make a thousand calculations about what might hold you today. They also make refusal socially expensive. If the invitations, jokes, announcements, photographs, and casual arrangements all happen inside the enclosure, the person outside it doesn’t become Spartacus. He becomes the awkward fellow who has to be emailed separately, assuming anyone remembers. Resistance is rarely cinematic. It is inconvenience, mild exclusion, and other people finding you faintly tiresome.

That’s the human cost of refusing. It is tempting to imagine ourselves standing athwart the geist of the zeit, saying “Stop,” or at least, “No, thanks,” while grateful citizens throw off their digital chains and take up the browser bookmark in solidarity. More likely, they will simply leave us out of things. Not maliciously. Not even consciously. If you are the only person who cannot be reached through Facebook, Instagram, WhatsApp, TikTok, Discord, or whatever walled garden has become obligatory this quarter, you are not a dissident to them. You are a logistical nuisance. You are not “resisting corporate internet” in the abstract. You are resisting social gravity.

None of this means the old web was pure. It was ugly, slow, full of dead links, pop-ups, malware, bad design, lunatics, hoaxes, and people who thought lime-green text on a black background was a legitimate aesthetic. Lots of personal blogs are, frankly, boring. Nostalgia should never be allowed to obscure reality. But the older web preserved something worth missing: the sense that the internet was a place you moved through, not a system that processed you. Its clumsiness left room for intention. You had to remember, return, and build your own routes.

That may be the most we can reasonably ask of the old fields now. They will not become the center of village life again. The train is too fast, too bright, too full of people we know. But the fields are still there, for those willing to get off and walk. A few badgers remain in the hedgerows, snuffling through old links, tending bookmarks, visiting sites by hand, and proving in their way that the enclosure was never complete.

Oaths

By in ai, technology

Our fathers took oaths as of old they took wives,
To have and to hold for the term of their lives,
But we take our oaths, as our whores, for our ease,
And a whore and a rogue may part when they please.

— Tom Brown, “Oaths”

Alan Jacobs has issued another encyclical on his AI usage. It turns out that the matter is more complicated than his earlier blanket rejections allowed. Claude, regrettably, is useful. Jacobs has bought a year of the Pro plan, though he assures us that renewal is not the default. Renewal will depend on “significant ongoing needs,” the relative corruption of Anthropic among the AI kaiju, and the possibility of reconciling his use with the energy and water demands of the company’s datacenters. No LLM will ever write so much as one word for him. It may, however, clean up dictated transcripts, sort and rename 1,400 PDFs, identify authors, search the web when necessary, and leave the results “very clean” and “extremely satisfying.” (There is one tiny comic grace note in the first paragraph: “(a)… (b)… and (b)…” The doubled “b” is almost too perfect. No LLM will ever collaborate with him, and by God, no LLM will fix his outline lettering either.)

Jacobs wants to take his oath as a wife, to have and to hold, for the term of his life. “No LLM will ever write so much as one word for me” has the ring of covenant. But the surrounding apparatus — the trial period, the reassessment, the acceptable uses, the ecological accounting, the mental image of Cousin Claude from Cullman to keep the machine from becoming an oracle — suggests that he fears he may be a rogue. Somewhere between “never one word” and “extremely satisfying,” the vow begins to feel the pressure of interpretation.

Readers of Jacobs over the years will recognize the shape. I say this with affectionate exasperation. Twitter used to summon the same cycle of attraction, renunciation, return under discipline, renewed disgust, and further renunciation. He would swear it off for excellent reasons, then reappear with a new framework for limited use, then conclude with embarrassment that the framework had failed. The pattern was not at all contemptible. In many ways it was admirable. Plenty of people are mastered by their habits and never write one honest sentence about it. Jacobs’s oddity is that he can’t seem to have a vice without writing its anthropology. If every ordinary technological decision becomes a spiritual crisis, one begins to wonder whether the crisis itself is part of the gratification. There is pleasure in renunciation, pleasure in relapse, pleasure in diagnostic clarity, pleasure in the newly drafted rule. The whole cycle can become a kind of moralized fidgeting.

For ordinary people, “I have a problem” is the beginning of action. For the intellectual, it is the beginning of a research program. Before the app can be deleted, one must consider repair, attention, embodiment, Christian humanism, technological enclosure, common life, convivial tools, perhaps Mumford, certainly Ellul, and, if time permits, the eschatological destiny of the human person under platform capitalism. The professorial reflex is: before I can decide what to do, I must understand the genealogy of my doing. What is the technological imaginary? What is the hidden anthropology of the platform? What does this reveal about late modernity, attention, embodiment, sacramentality, political economy, and the shape of the common good?

In the Buddhist parable of the poisoned arrow, the wounded man refuses to let the physician remove the arrow until he knows who shot it, what caste the man belonged to, what kind of bow was used, what sort of feathers were on the shaft, and so on. Of course, by the time the inquiry is complete, he is dead. Jacobs has been shot by Twitter, or Claude, or the internet, and instead of pulling the arrow out, he begins assembling a seminar: Mumford on technique, Ellul on technological society, Illich on tools, Augustine on disordered loves, Doctorow on enshittification, perhaps a short excursus on tikkun olam, followed by a cautiously hopeful note about decentralized protocols and the redemocratization of the internet. At some point, “What should I do about my Twitter/AI habit?” has been smothered under so many layers of diagnosis that the obvious answer — “stop looking at Twitter/AI” — seems almost vulgar.

This is where a grounding in less theoretical forms of work has its uses. In a warehouse or a small business, abstraction has a shorter leash. The pallet either moves or it does not. The order ships or it does not. The customer gets the right item or writes back annoyed. Rent, postage, inventory, fatigue, floor space — these things have a way of disciplining the mind. I do not need to reimagine the warehouse every time a pallet is in the wrong place. Sometimes the answer is not a theory of space, labor, and late capitalism. Sometimes the answer is: move the pallet.

This doesn’t make practical people immune to self-deception. They procrastinate, rationalize, scheme, and backslide like everyone else. But the work itself is less patient with ornamental thought. A theory that does not cash out in action soon starts looking expensive. Academia, by contrast, can reward the elaboration of the problem independent of its solution. In fact, the elaboration often is the product. So the professorial vice is not laziness in the ordinary sense. It’s productive avoidance. You are working, reading, connecting, citing, historicizing, complicating — and yet somehow the arrow remains lodged exactly where it was.

Someone who is genuinely indifferent to a temptation does not usually stage elaborate demonstrations of indifference. He just ignores the thing. The theatrical proof suggests that the thing has already taken up residence in the imagination. The act of mastery becomes another way of keeping the object close. “I shall use the infernal convenience-machine, but only as valet, clerk, indexer, recipe-card, road atlas, and moral eunuch. It shall never become my co-author.” Maybe true. But anyone issuing regular communiqués about the chastity of the arrangement probably knows there is some danger in the room. As the meme-joke has it, his “My Relationship with Claude is Ethical and Chaste” t-shirt has a lot of people asking questions already answered by his shirt.

Aesthetic Prion Disease

By in ai, technology, the statusphere, writing

Sam Kriss:

I’ve had enough. Which is why, from here on out, if I see you passing off an AI output as your own work, I am going to kill you. I will find out where you live and bash your head in with a crowbar, until the brains you decided not to use are dripping down the walls. This is not a literary device. This is not a comic bit. This is a highly credible real-world threat to do physical harm, punishable in America under 18 USC § 875(c) and in the UK under section 16 of the Offences Against the Person Act.

Kriss goes on to argue that beauty and insight have already been lost to the machines. If you attempt to write in that mode now, you are trespassing on their territory. What remains inviolably human is not judgment, taste, memory, craft, or responsibility, but degradation: real writers are “sexual perverts,” “bitter and frustrated little gremlins,” “rapidly obsolescing apes” driven by “dumb lust and jealousy.” The machine may have won beauty; we retain “death threats, racist tirades, and pornographic fantasies.” Well. With friends of the human spirit like this, the robots can afford to take the afternoon off.

This is Romanticism minus the aspiration. The old Romantic complaint was that Enlightenment reason had made the world too neat, too dry, too mechanical. Man had to be rescued through passion, imagination, nature, sublimity, eroticism, madness, rebellion. Fittingly for an age of undead IP franchises, Kriss offers the rebooted and degraded version: if machines can imitate poignancy and insight, then the human writer must prove himself through spite, obscenity, and menace.

The other thread in the essay is less Romantic than professional: the guild-reflex. Kriss is a man with postgraduate training in Critical Theory, a long record of publication, and a livelihood bound up with being recognized as a writer (at least until his wealthy parents die and leave him their “massive house“). Now automation has wandered onto his turf. His fury is territorial. This is the prose equivalent of lifting one’s leg.

It’s true that much AI writing is bad in precisely the way he says it is bad: maudlin, frictionless, padded, eager to please, always reaching for poignancy without earning it. But much human writing has always been bad in the same way. The machine did not invent false profundity, stale metaphor, obscure nonsense, or the thrill people get from mistaking vagueness for depth. Hegel, Heidegger and Zizek did not need ChatGPT. Generations of poets, mystics, academics, commencement speakers, essayists, and newsletter moralists have managed to produce deepities under their own power.

What AI changes is not the existence of bad writing, but access to its surface competence. A person without the usual training, leisure, inborn talent, or literary inheritance can now ask for help organizing an argument, finding a phrase, testing a paragraph, or sanding down a clumsy sentence. If you need that sort of help, the implication runs, perhaps you were never meant to be here. Go back to assembling widgets, sweeping floors, driving Uber, or whatever else the uninitiated do with their days. The title of Writer is not yours to share.

Like it or not, AI is now another active agent in the literary ecology. It is not some alien corruption descending on an otherwise pure republic of letters. It is another producer of verbal atmosphere, though an unusually powerful one because it can generate enormous quantities of plausible prose on demand. The important point is that it imitates the existing discourse, then feeds its own imitation back into the discourse, where humans imitate the imitation, which machines then ingest again. Aesthetic prion disease, basically.

And the funny part is that human beings have always been machine-like in groups. Give any scene enough time and its members converge on the same rhythm, vocabulary, in-jokes, pieties, insults, and other verbal mating calls. Most “original voices” are partly just talented instances of a local dialect. AI makes this more obvious, and maybe more depressing, because it shows how much of our supposedly personal expression was already patterned.

People talk as if AI prose will simply flatten everything forever, but annoyance is one of the great engines of aesthetic change. The more dominant the median machine cadence becomes, the more valuable it becomes to sound unlike it. Some people will overcorrect into conspicuous difficulty: God help us if more people try to write Finnegans Wake. More jagged syntax, more private references, more deliberate ugliness, more broken forms, more “you can’t autocomplete this, you bastard.” That may produce some interesting work but will certainly produce a lot of unreadable peacocking.

Evolution proceeds. A new predator appears; the prey get faster, spikier, more camouflaged, or better at tasting like poison. The bot-species enters the ecosystem, and writers who care about style develop new defenses. Some responses will be hysterical. Some will be snobbish. Some will be productive. The worst writers will imitate the machine. The vainest writers will perform anti-machine authenticity. The better ones will absorb the lesson and keep moving.

1 2 3 4›»