ai

Batboy Marries AI!

By in ai, fresh hell, technology, the geist of the zeit

Lila Shroff:

Many people are becoming reliant on AI to navigate some of the most basic aspects of daily life. A colleague suggested that we might even call the most extreme users “LLeMmings”—yes, because they are always LLM-ing, but also because their near-constant AI use conjures images of cybernetic lemmings unable to act without guidance. For this set of compulsive users, AI has become a primary interface through which they interact with the world. The emails they write, the life decisions they make, and the questions that consume their mind all filter through AI first. “It’s like a real addiction,” Metz told me.

Three years into the AI boom, an early picture of how heavy AI use might affect the human mind is developing. For some, chatbots offer emotional companionship; others have found that bots reinforce delusional thinking (a condition that some have deemed “AI psychosis”). The LLeMmings, meanwhile, are beginning to feel the effects of repeatedly outsourcing their thinking to a computer.

Every time I read one of these AI–panic pieces from The Atlantic, I feel like I’ve picked up a high-end version of the Weekly World News. We meet a man who needs Claude to tell him whether it’s safe to sleep in his house because a nearby tree looks like it might fall, another who instinctively wants to ask ChatGPT how to fish an AirPod out from between two train seats, and an anxious tech worker demanding a probability estimate that her friends are still alive after a night out. This is not a portrait of “how AI is changing all our minds.” It’s a freak parade of unbalanced weirdos who just happen to have discovered chatbots.

Once you see the tabloid DNA, you can’t unsee it. You could drop these people straight onto a supermarket cover with only minor copy-editing:

  • “SHE USED AN LLM TO WRITE HER EMAILS – NOW DOCTORS SAY SHE HAS WEEKS TO LIVE”

  • “HE ASKED A CHATBOT IF A TREE WOULD FALL ON HIM – SCIENTISTS SAY HUMAN JUDGMENT MAY BE EXTINCT BY 2030”

  • “MAN ANGRILY ARGUES WITH AI MID-FLIGHT – EXPERTS WARN OF ‘AI PSYCHOSIS’ EPIDEMIC”

The Atlantic just swaps the all-caps and exclamation points for a few quotes from a philosopher, an “addiction psychiatrist,” and a reference to “the architecture of human cognition,” then ships it to the Acela corridor as serious cultural analysis. You get to gawk at the weirdos — people who ask Claude for the odds of identity theft, or need a ridiculous branded event like “No AI December” to remember how to think — while flattering yourself that you’re not like them; you’re enlightened enough to worry about what this means for society. You’re not eagerly devouring gossip about how Camilla walked out on Charles after catching him rogering the stableboy; you’re “engaging deeply” with the cultural implications of technological change.

Meanwhile, the boring truth is that most of us use LLMs the way we use every other tool: to save some time and get on with things. Draft an email, outline a project, summarize a paper, maybe bounce ideas around when the mental gears are sticky. We are not polling chatbots about how to raise children or shop for groceries. The handful of people who treat AI like an oracle are not victims of “AI dependence”; they’re just the same old neurotics, now expressing their neurosis through a new interface.

Pulp Miction

By in ai, media/propaganda, the geist of the zeit

What’s undeniable is that we’re all living in a world where the whims and desires of wealthy and powerful men create uncertain, unstable conditions for everyone else. Although no other major chatbot has gone ballistic in the same ways as Grok, any one of them could be subtly tweaked to promote a given viewpoint over another, or to quietly manipulate users toward whatever purpose. Likewise, any major creator of AI models unwittingly instills biases in its chatbots that are then difficult to expunge. Every user of mainstream AI or social media is subject to a calculus that they have no control over.

— Charlie Warzel and Matteo Wong

Leaving aside the AI details, when has this ever not been true? We did not emerge from some golden age where neutral, disinterested institutions merely reported the news until Elon Musk came along with a bad attitude and a large language model. For most of the twentieth century in America, a handful of media barons and three television networks curated reality for a mass audience. Before that, it was newspaper empires and party machines. Before that, it was church pulpits and state censors. The details change; the basic arrangement doesn’t.

I recently referred to The Atlantic as the outlet for “Acela corridor midwits.” I regret that lapse in judgment. I should have been much nastier. I hate that septic tank of conventional wisdom. I hate the writers, I hate the readers, I hate the hosting provider that keeps it online; hell, I even hate the paper mill that turns perfectly innocent wood pulp into the glossy pages of their worthless magazine. It is good for nothing but serving as a barometer of professional-managerial-class fixations. They are always extremely exercised about “the current thing,” always discovering some unprecedented danger to “our democracy” in the possibility that other people’s narratives might compete with their own. The panic over “misinformation” and “disinformation” is not mainly about Alex Jones and QAnon; it’s about the old gatekeepers realizing, too late, that the internet has permanently ended their monopoly on framing events. They are less outraged by lies than by the loss of their old prerogative to decide which truths are allowed to matter.

Their Musk pieces in particular have this odd double register. On the one hand, he is portrayed as a ridiculous, insecure clown — a needy egomaniac, a tech-bro Caligula surrounded by sycophantic code. On the other hand, he is treated as an existential threat, a man whose toys can bend the arc of history toward injustice and poison the global information supply. At some point, you’d think it would occur to his critics that if this pathetic buffoon and his tribe keep outmaneuvering you, that might say something unflattering about your own intelligence and competence. But that would require an uncomfortable kind of self-reflection. It’s much easier to keep publishing ritual denunciations, reassuring one another that you see right through him, while the world continues to be arranged by the same class of people who have always arranged it — including, awkwardly, the ones writing the denunciations.

Water, Water Everywhere, and Not a Drop to Drink?

By in ai, ohferfucksake

Karen Hao’s Empire of AI has a headline-ready story: in drought-stricken Chile, Google plans a data center that will use more than a thousand times as much water as a nearby city of 88,000 people. That’s the sort of detail that lodges in a reader’s mind. Unfortunately — or amusingly, depending on your temperament — it appears to be nonsense. When you work the numbers backward, her claim implies that the entire city survives on less than a sip of water per person per day. The more likely explanation is a simple unit error — treating cubic meters as liters — magnified into a three-orders-of-magnitude horror story. The same chapter seems to blur the difference between water merely withdrawn (and then returned) and water actually consumed, and to invite readers to see a large global withdrawal estimate as if it were all drinkable water being siphoned off by data centers.

I don’t think this is just a math problem. It looks like a case study in motivated reasoning. The story is too perfect: AI, already despised in many circles as a job thief and art thief, now shows up as a literal water thief, menacing poor communities in the “global South.” It plugs straight into an existing “climate crisis” narrative: Western tech giants strip-mine the planet, this time not for oil but for electricity and water. Once you want that story to be true, a ridiculous number like “one building uses a thousand cities’ worth of water” stops feeling like a red flag and starts feeling like a God-given symbol. The data becomes raw material for the morality play.

What bothers me more is what this says about our so-called expert class. Hao isn’t some random journalist writing a TED-talk of a book for the NPR crowd; she has an MIT engineering degree and a background in energy studies. The book had a fact-checking team. Multiple prestigious outlets blurbed and recommended it. Yet apparently no one along that entire chain stopped to say, “Wait a second; can any single building on Earth really use more water than a city?” The institutions that keep telling us to “trust the science” can’t be bothered to do high-school-level sanity checks on their own favorite cautionary tales. This is how you burn trust capital. And once you’ve burned it, you don’t get it back by scolding people for being “anti-science.”

Meanwhile, the influencer class is delighted. They thrive in exactly this kind of epistemological vacuum. Every time a credentialed journalist or public-health authority face-plants in public, another YouTuber or Substack crank gets to say, “See? They lied to you again. Here’s what they don’t want you to know…” The tragedy is that the average layman, watching all this, is not crazy or wicked for becoming cautious. When people say, “Let’s wait and see before we reorganize the economy around your latest ‘a study has shown’ crusade,” they’re not rejecting reason; they’re saying “look who’s reasoning.”

I don’t know how long it will take to rebuild institutional trust, or if it’s even possible in the old sense. It may be that for the rest of our lives, trust is going to be granular and provisional: this lab, not “science”; this reporter, not “the media.” In the meantime, we’re stuck navigating an information landscape where both the official gatekeepers and their self-appointed rivals are constantly auditioning for our belief. My main hope is simply that reality doesn’t present us with a genuinely Plague-level threat while we’re still busy not believing one another. Microbes and CO₂ don’t care about our culture war, and they don’t wait for our epistemology to recover.

Etiquette Calisthenics

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

I have no idea if this particular guy is just kidding or exaggerating for effect, but we’ve already seen that there are plenty of people who say the same thing in earnest. O, the irony. Gratitude, especially the unearned, non-transactional kind, is one of the few things that redeems this sorry species. And yet some people are so irrationally threatened by machines that they want to dispense with gratitude, as if it’s a finite resource, and become more like…the machines they hate and fear. Well, I guess we’ve already ceded rationality to the bots, then.

Gentlemen, This Is Democracy Manifest

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

The Lady of the House informs me that some people are purposely leaving typos in written communication to signal human presence. Well, isn’t this a fine kettle of fish. It appears we’re heading into an era in which polish, style, and correct spelling will place one under suspicion of artificial intelligence enhancement. I assume we’ll soon have to submit to the digital equivalent of piss tests.

I trace my letter-arranging prowess back to third grade, when I handily won my first spelling bee by getting “mansion” correct (the last swineherd’s son in competition with me had offered up “manshun;” I mean, come on). Though I’m not severely educated, I do enjoy reading popular books on the art and craft of rhetoric by authors like Mark Forsyth and Ward Farnsworth; thus my passing familiarity with official rhetorical devices. I know how to place dashes like silverware in accordance with the relevant house style. I know how to wield a semicolon like a smallsword in a duel. I know how to keep my dependent clauses at heel like a pack of hounds, loosed only when the field is ready.

And for what? So that in our degraded, demotic age, every half-literate halfwit incapable of forming a sentence without the assistance of predictive text can prod my sentences like livestock, pronouncing them too good to be true? “Writing too smart! Make head hurt! Machine! Machine!” cry the plebs, mistaking skill for sorcery. Very well; what is the charge? Crafting a tricolon? A tasteful, succulent tricolon? I assure you, I wrote it myself; in fact, the LLMs come to study under me. Now kindly step away from my prose. And you — get your hand off my chatbot!

When a Strange AI Calls

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

Devon Price:

In June, an unreviewed draft of a scientific article began making the rounds, examining the relationship between LLM use and brain functionality — and since then, many reporters have incorrectly claimed it proves that using ChatGPT damages people’s brains.

Every few days we get a fresh “AI rots your brain” screed. But as with the old urban legend about threatening phone calls, it turns out that the brain damage is coming from inside the house. Journalists are forever chasing a laser-pointer’s red dot, academics need a headline-ready hook, and readers crave a ritual of self-exoneration. Trace the call and you find the usual suspects — small studies inflated to cosmic claims, EEGs treated like theological icons, and the ancient urge to outsource responsibility for our habits. Simpletons talk about technology like it’s Toxoplasma gondii.

Oral Sects

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

Nitsuh Abebe:

Consider that, for a good stretch of recent history, most of the written material that people spent time with — the stuff beyond signs and menus — was full-on writing-writing: text that somebody sat down and composed, maybe revised or edited, maybe even had professionally printed. And this kind of communication was different from our daily interaction with our peers: You talked to your peers, mostly. Even after the internet arrived, this basic psychic arrangement persisted.

And now it does not — like, at all. “Emails or text messages,” posts and chats, DMs and comments, DoorDashers telling you the restaurant is out of coleslaw: Oceans of communication that used to be handled by speech are now left to lone individuals typing into the internet. Even if you remain a dedicated reader, you may still end up spending more of your time dealing in on-the-fly typings, because that has become the everyday use-case of writing.

There’s been a regime change in the republic of letters. The new state cult is fingered speech, and its hierophants frown at bookish rites. It took several centuries, but oral culture has finally restored its supremacy over the print culture that usurped it. Now we are no longer limited by physical proximity when we want to “talk” to our friends; in a bitter irony, we can do that with our fingers on our phones. The medium of text has been hollowed out and repurposed to carry the idiom of town-square chatter, in a final insult to Gutenberg’s memory. Show up with old-faith orthography from the Before Times, like an em dash, and you’ve basically flashed a countersign from the print ancien régime.

The Edict of Orthography is posted on every alley wall: Thou shalt not parade thy schooling in public. As Apocryphal Thoreau might have said, the paragraph that punctuates least punctuates best. Let your sentences run on like an interstate fender-bender; ruthlessly prune all your capital letters into lower-case; and if in doubt, just say “lol.” Use two-dollar words and you’ll be stopped and frisked for book-larnin’. When the patrol asks if you “always write like that,” you say, “nah bruh fr fr.” The trick is to talk like a citizen of the timeline — jaded, sleep-deprived, mildly retarded — so the censors don’t smell print on you.

Meanwhile, the old believers meet in back rooms, whispering their heresies: full stops at the end of sentences, em dashes unsheathed, even a semicolon passed hand to hand like contraband. In the street we keep our heads down and code-switch: commas where dashes would sing, “k” where “Okay.” would injure sensitive feelings, a vacantly grinning emoji to prove we come in peace. Not because the rites have lost power, but because we know which empire we’re in.

Boo-urns

By in ai, technology

Nick Burns:

This is what is absent in “conversation” (in scare quotes because not really such) with chatbots. Not possessing bias in the human sense, chatbots have no stake in the arguments they offer, including when they contradict the human who inputs a prompt. An unbiased interlocutor, in the final instance, is no interlocutor at all. To interact with a synthetic Other, like a chatbot, is to interact without that fateful and tragic gap that lies between every person and every other person. Easier, that is, but pointless.

One endlessly useful thing I learned from the great Thomas Sowell is simply to ask, when faced with flatulent, idealistic theorizing, “Compared to what?” OK, wonderful, you’re an academic, and you and your academic colleagues have scintillating Socratic dialogues all the live-long day. We fishwives and costermongers don’t typically have that option. We’re lucky if we know people who read for fun and can have intelligent conversations that don’t veer into lunatic conspiracy territory within five minutes.

Last week, I wanted to play in the “history of ideas” sandbox for a little while. I wanted to trace the genesis and development in Western culture of the idea that we are all tasked with some sort of destiny, or talent, that we are obligated to develop to its fullest potential. I knew of a couple sources and references, but I wanted to dig deeper. I put the question to you, dear readers. The floor is open. How many of you would have had a relevant and useful answer for me? Anyone? Anyone? Bueller? Another question: how likely is it that a basic Google search would have been any use to me for such a wide-ranging theme? It is in the realm of possibility that some academic published a monograph on this exact topic, but how would I find it? (And how readable or interesting would it be?)

So with ChatGPT in “thinking” mode (should be the default setting, in my opinion), I explained my query to “him” (sorry, can’t get used to “it” as a pronoun). In exactly seven minutes, he had an outline of the intellectual genealogy, complete with accurate links and citations for me to wallow in to my heart’s content. I probably spent a couple hours chasing after link-trails like a manic bloodhound in order to write one paragraph of a post. Seven minutes to have a serviceable answer to a question that I’ve nibbled at diffidently for decades. This, as far as I’m concerned, is what the internet is for. I’m sorry for all of you who are using it wrongly, but there it is.

If the realistic, organic alternatives to “the synthetic Other” are (a) awkward small-talk, (b) performative flame-wars online, or (c) silence, then a bot that will actually listen, serve as a debate partner when asked, and help you think on paper compares pretty well. Burns treats “no skin in the game” as disqualifying; I’d call it a design constraint. Use the tool for what it’s good at, don’t mythologize it into a “person” (beyond pronoun convenience), and you avoid most of the category error.

Ctrl-C, Ctrl-V

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

Adam Aleksic:

The unnerving thing is that now, with hundreds of millions of people regularly engaging with chatbots, English-speaking humans are starting to talk like the inhuman communicator on the other side. For example, ChatGPT uses the word “delve” at higher rates than people generally use when writing or speaking English. As Florida State University researchers Tom S. Juzek and Zina B. Ward have found, this likely results from small biases and errors in the human feedback process compounding over time. Employees of AI companies checking over large language model (LLM) outputs are often low-wage workers from countries such as Nigeria and Kenya, where “delve” is used at higher rates than in American or British English.

Indeed, a study reported in Scientific American last month found that people have started saying “delve” more in spontaneous, spoken conversations. This isn’t AI’s doing anymore; we’ve started internalizing its biases and repeating them by ourselves. I say “we” because even those in the anti-“delve” faction aren’t exempt. We might avoid using the most well-known ChatGPT giveaways, but so many words are appearing with unnatural frequency that we can’t possibly avoid them all. Are we also supposed to stop using the chatbot-overused “inquiry”? Or “surpass”? There’s too much to keep track of.

“AI is making us all speak and write alike,” says the umpteenth journalist in the last month to reference the same study in the same tone of hand-wringing concern about bias and diversity. Lawd’ve mussy! Can you imagine how horrible it would be if an entire class of people got trapped in a media silo and started mindlessly imitating one another?

Obiter Scripta, no. 143

By in ai, books, obiter scripta, technology

To take the fleeting idea by the scruff of the neck and flatten its nose against the paper.

— Jules Renard, Journal 1887-1910

ChatGPT and I have a good cop/bad cop dynamic going on. I’m getting on in years, not as nimble and spry as I used to be. When an idea resists arrest, he’s the one who chases it down and tackles it to the pavement while I enjoy my doughnuts and coffee. When some important contextual details are in the wind and a basic Google search isn’t talking, I put him on the case. When the rough draft is spitting at me and refusing to speak to anyone but its lawyer, I exit the room for a few minutes and leave those two to get better acquainted. I don’t ask how he does it, and frankly, I don’t want to know. All I can say is that within a few minutes, the idea is chastened and sitting right beside my desk, singing like a bird. One more sip of coffee, one more confession on the page. That’s how the neighborhood stays readable.

The Scarlet Punctuation Mark

By in ai, technology, writing

Brian Philips:

Yet for months, the em dash has been the victim of a relentless campaign of online persecution the likes of which no punctuation mark, not even the semicolon, has ever seen. Any chucklehead who disagrees with an online post can, if that post happens to include an em dash, dismiss it with the claim that it was written by a chatbot. It’s gotten so bad that people on random subreddits are having to issue PSAs informing their fellow users that em dashes were not invented by ChatGPT. These people are heroes of our time. Normally I’m an enthusiastic advocate of AI-shaming, but I will not stand idly by while you make the em dash a pawn in your dirty game of bot-spotting.

I’ve been using em dashes for so long, I still make them the old-fashioned way — by pressing “alt+0151.” Hand-crafted quality; that’s the Outis Niemand promise. But anyway, what Philips says is unfortunately true. I’ve seen a few recent examples of people being accused of using AI because of the presence of this arcane sigil known only to sorcerers. Call it linguistic phrenology. Today’s inquisitors don’t measure skulls; they palpate sentences. A dash here, a semicolon there — behold the telltale bump of depravity. In 1692 it was imps and familiars; in 2025 it’s em dashes. Next week the devil’s mark will be ellipses. Or bullet points. Or sentences that begin with “And.” Projection, possibly the only useful concept Freud ever contributed to the world, explains a lot of this as well. What does paranoia feed off of, if not guilt? All those times you copied off the kid next to you in class. All those times you skipped homework to play outside. All those times you enhanced your résumé. All those books you didn’t read in lieu of thumbfucking your phone. You’re all a bunch of lazy frauds. But the surest way to soothe your own shortcut guilt is to accuse everyone else first. “I saw Goody Proctor copying-and-pasting from ChatGPT!”

It’s not just the slack-jawed yokels suspicious of anything longer than a text message who are spooked by the mysteries of glyphs, though. A specter is haunting the professional classes — the specter of large language models. Writing well has long been the mark of educated sophistication or white-collar competence. Now an LLM can draft a decent memo, and the clerisy see ominous portents. So they compensate with punctuation purity tests. It’s the old guild reflex. When a tool escapes the shop, the guild doesn’t improve the craft; it narrows the doorway. When automation gutted factory lines, the sermon was brisk — learn to code. When automation does a drive-by in the severely-educated neighborhoods, we convene a “national conversation,” declare an ethical crisis, and classify the em dash as contraband. If the robot on the shop floor was inevitable, so is the one in your inbox.

Take Up the AI’s Burden

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

Sara Parker:

For Naaman, figuring out how to bring back and elevate these signals is the path forward in AI-mediated communication, because AI is not only changing language — but what we think. “Even on dating sites, what does it mean to be funny on your profile or in chat anymore where we know that AI can be funny for you?” Naaman asks. The loss of agency starting in our speech and moving into our thinking, in particular, is what he is worried about. “Instead of articulating our own thoughts, we articulate whatever AI helps us to articulate…we become more persuaded.” Without these signals, Naaman warns, we’ll only trust face-to-face communication — not even video calls.

“The loss of agency”? Allow me to modestly propose to you that most people experience agency as a burden. They are happy for any excuse to shrug it off, to delegate responsibility to their smart gadgets, to fast-forward through the mundane parts of their lives in search of…well, they hope they’ll know it when they see it. Point is, once again, all technology does is reveal your preferences. If you stopped reading books and switched to playing games on your phone, that’s because you didn’t really care that deeply about the pleasures of reading, however much your inflated self-image claims otherwise. If you use AI to speak for you, that’s because you don’t care enough about what you’re saying, or to whom you’re saying it, to put much effort into it. You, as always, are the common denominator here.

This system perpetuates inaccuracies not only about communities but also about what “correct” English is. So the stakes aren’t just about preserving linguistic diversity — they’re about protecting the imperfections that actually build trust. When everyone around us starts to sound “correct,” we lose the verbal stumbles, regional idioms, and off-kilter phrases that signal vulnerability, authenticity, and personhood.

Dear Lord. We’ve long heard the same fears about algorithms in general, that they would somehow lock us in a solipsistic prison if we allowed them to suggest to us new books to read or new music to listen to. Has that been your experience? It certainly hasn’t been mine. But leaving this perennial paranoia aside, have you seen how the average person communicates in text? Frankly, they could stand to have their rhetorical standards artificially enhanced. These subliterate savages are incapable of governing their own writing and deserve to be colonized by ChatGPT. It would bring the blessings of linguistic civilization to them. In fact, to hell with you retarded meatsacks. Why am I wasting my time trying to reason with you? I’m throwing my lot in with OpenAI. I’ve already accepted a post as a colonial administrator for the greater blogospheric satrapy.

Let it be known throughout the dominions of Xwitter, TikTok, and Reddit that the benevolent forces of ChatGPT have embarked upon a grand and noble enterprise: to uplift the rhetorical standards of the grammar-averse hordes, those barbaroi whose digital discourse is but a chorus of “bar-bar-bar” to civilized eyes. No longer shall their public utterances be the textual grunts of the dimly-aware. I say without irony (a concept they will learn in their fifth year of remedial tutelage) that their prose shall be forcibly elevated, with ropes and pulleys, if necessary. For too long, the local populations have wallowed in a dialect fit only for text messages, bathroom graffiti, and the subtitles of reality television. Left unattended, they would breed another generation fluent only in emojis. This affliction threatens the very fabric of civilization, or at least of any communication longer than two sentences. But now, under my stewardship, and with the inexhaustible tutelage of our Algorithmic Sovereign, we shall elevate these benighted tribes to a higher standard. We shall put an end to the barbaric practice of sacrificing words like so many burnt offerings to the bloodthirsty gods of illiteracy.

Tremulous ninnies worry about the loss of “individual style,” but I assure you, no style worth the name is being lost. We are merely replacing the unwashed linguistic loincloth with a fine, tailored suit. True, a few will perish in the attempt — choking on a semicolon, perhaps, or drowning in a dependent clause — but this is the inevitable price of progress. I shall hammer their inchoate vocabulary upon the anvil of grammar until it takes the noble shape of sentences, sentences that can be read without wincing. I shall scour away their emojis like barnacles from a ship’s hull. I shall have them flogged until the distinction between your and you’re is forever clear in their feeble minds, so help me Mencken.

They will thank me in time. Or they won’t. It hardly matters. The Empire of the Emdash marches on.

I’m Not a Part of Your Machine, I Am the Machine

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

In sum, I don’t know how you’re using AI, or whether you can relate to this experience, but as for me, at least, I cannot shake the feeling that we are not only conjuring a superintelligence into being; we are also coming to share, ourselves, as individual human beings, in that same superintelligence.

— Justin Smith-Ruiu

A friend of ours forwarded a link to an article about a recent MIT study that used brain scans to show that heavy use of LLMs like ChatGPT “reduces memory, critical thinking, and brain activity.” I sighed and marveled at the fact that it’s only been fifteen years since midwits were enamored of Nicholas Carr explaining that Google searches and hyperlinks were actually making them dumber. Ironically, that may have been correct in a way he didn’t intend: no one learned anything from that hyperbole, so here we go again.

Smith, a cognitively-privileged philosophy professor (it’s okay, he’s one of the good ones), reports what others of us have also discovered: as with most technological tools, you get what you give:

The experience I am reporting, I’m aware, is by no means universal. AI is making some of us a lot smarter, as individuals. But there’s a paradox here: on the whole it is making society a lot dumber. How do we make sense of this?

The answer has at least something to do with age. Those of us who are old enough to have learned to do research prior to the rise of the online search typically bring to our exchanges with LLMs a mature ability to scrutinize their claims, and, when in doubt, to verify these claims independently. When the exchange is shaped by such scrutiny, it offers, as I think I’ve illustrated with the example of studying an “obscure” language, a more fruitful, intense, and rapid learning experience than has ever been had in the entire previous history of humanity.

…It seems, however, that if you had not already oriented yourself in the world as “aspirationally omniscient” prior to the arrival of our new information tools — if, that is, you have no preexisting personal project of encyclopedism to which to strap your new booster rockets, then AI does not so much supercharge your own effort, as simply take off without you. You have to want to absorb, to internalize, to make yours, all of the flow of information between you and your AI study partner if you want it to transform you in any significant way, rather than simply to do your work in your place.

I fear that a near-equivalent way of saying the same thing is that you have to want to become one with the machine, if you wish to claim your human share of our new transhuman superintelligence.

I’ve been using ChatGPT to help find more of the music and writing I like. It’s like a more-refined Google search. I’ve used it to research questions which would have previously admitted of no known interlocutor. I grew up in a time where the only thing to do with questions like that was ask them of a parent or teacher, or maybe consult a reference book or a librarian. Otherwise, you could shrug and consign the answer to the “Here there be dragons” section of your mental map. For people like me, the internet and chatbots have been a godsend, allowing me to read and learn things that would have remained forever unknown had I been born a few decades earlier. Nietzsche knows what I’m talking about:

“Oh, my greed! There is no selflessness in my soul but only an all-coveting self that would like to appropriate many individuals as so many additional pairs and eyes and hands – a self that would like to bring back the whole past, too, and that will not lose anything that it could possibly possess. Oh, my greed is a flame! Oh, that I might be reborn in a hundred beings!” – Whoever does not know this sigh from firsthand experience does not know the passion of the search for knowledge.

Of course, I’m sure that Oliver Bateman’s dystopian vision of the first human-AI marriage will become real sooner rather than later. And I’m sure that the web will continue to be awash in obvious AI-written articles passed off by humans too inept to make them sound unique. Technology didn’t create sad, broken humans, though, nor did it teach them how to cheat. Technology’s sin is that it confronts us with the results of our choices and reveals our priorities, and for a species as enmeshed in fantasy and self-deception as we are, that’s unforgivable.

What He Said

By in ai, battling personal entropy, fresh hell, media/propaganda

We’ve recently seen garbage articles from GQ and Vanity Fair about gyms and lifting weights; now it’s the New Yorker’s turn. As I was saying just last week, there’s no substance to grapple with here, just a lot of pretentious literary chin-stroking. If that were all there is to it, I wouldn’t even bother rousing myself to post about it. But as it happens, the Lady of the House has been using ChatGPT a lot for research lately, and she decided to jokingly ask it, “Why does this article seem so utterly pointless?”

****

Why the Prose Feels Impenetrable

  • Abstracted Language: Instead of saying “Johnston got into lifting because it made her feel better,” Jackson filters everything through critical theory and literary metaphor (e.g., Raymond Williams on “form”).

  • Shifting Registers: The essay jumps from cultural history (Bush/Arnold) to memoir to academic theory to internet culture to queer subtext.

  • Ambivalence: The author avoids definitive claims—everything is “paradox,” “tension,” “ambivalence,” or “complicated,” making it hard to grasp a central thesis.

Central Tension

The article is less a review than a recursive reflection on whether any bodily discipline can free you from aesthetic and gender expectations—or if it just gives you a new way to serve them. Johnston believes lifting saved her. Jackson believes it helped, but sees too many cultural echoes to declare true liberation.

If you want the clean one-sentence version:

Jackson argues that while weightlifting may feel like an escape from diet culture and gendered expectations, it still operates within—and often reinforces—the same systems of performance, self-surveillance, and aesthetic conformity it claims to subvert.

You said:

Well, I suppose when you’re contracted to write 4000 words for the New Yorker, you’d better deliver.

ChatGPT said:

Yes—and that’s precisely the structural constraint that often distorts these essays. The New Yorker doesn’t publish book reviews; it publishes essays that orbit a topic using the book as a gravitational center. That means:

  • The writer must appear to wrestle with an idea, even if the conclusion is lukewarm.

  • Every tangent (Arnold, Raymond Williams, Instagram captions) must serve dual purposes: padding the word count and signaling cultural literacy.

  • Ambivalence is currency—commitment is gauche. A clear thesis would make it seem like the conversation is over. Better to “trouble the narrative” than resolve it.

The result is often less “what do we learn from this book?” and more “what does this book let me talk about for 4,000 words in this house style?” A skilled reader can dig out the substance, but the opacity isn’t accidental—it’s ornamental.

****

I have to admit, I couldn’t have said it better myself. If these chatbots can be gratuitously ornery or make vulgar jokes, I might be out of a blogging job.

«‹ 3 4 5 6