Like you, I'm sick to the back teeth of talking about AI. Like you, I keep getting dragged into AI discussions. Unlike you‡, I spent the summer writing a book about why I'm sick of writing about AI⹋, which Farrar, Straus and Giroux will publish in 2026.
‡probably
⹋"The Reverse Centaur's Guide to AI"
--
If you'd like an essay-formatted version of this thread to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
A week ago, I turned that book into a speech, which I delivered as the annual Nordlander Memorial Lecture at Cornell, where I'm an AD White Professor-at-Large. This was my first-ever speech about AI and I wasn't sure how it would go over, but thankfully, it went great and sparked a lively Q&A.
One of those questions came from a young man who said something like "So, you're saying a third of the stock market is tied up in seven AI companies that have no way to become profitable and that this is a bubble that's going to burst and take the whole economy with it?"
So I re-iterated the book's thesis: that the AI bubble is driven by monopolists who've conquered their markets and have no more growth potential, who are desperate to convince investors that they can continue to grow by moving into some other sector, e.g. "pivot to video," crypto, blockchain, NFTs, AI, and now "super-intelligence."
Further: the topline growth that AI companies are selling comes from replacing most workers with AI, and re-tasking the surviving workers as AI babysitters ("humans in the loop"), which won't work. Finally: AI *cannot* do your job, but an AI salesman can *100%* convince your boss to fire you and replace you with an AI that *can't* do your job.
When the bubble bursts, the money-hemorrhaging "foundation models" will be shut off and we'll lose the AI that can't do your job, *and* you will be long gone, retrained or retired or "discouraged" and out of the labor market, and *no one* will do your job. AI is the asbestos we are shoveling into the walls of our society and our descendants will be digging it out for generations:
The only thing (I said) that we can do about this is to puncture the AI bubble *as soon as possible*, to halt this before it progresses any further and to head off the accumulation of social and economic debt. To do that, we have to take aim at the *material basis* for the AI bubble (creating a growth story by claiming that defective AI can do your job).
"OK," the young man said, "but what can we *do* about the crash?" He was clearly very worried.
"I don't think there's anything we can do about that. I think it's already locked in. I mean, maybe if we had a different government, they'd fund a jobs guarantee to pull us out of it, but I don't think Trump'll do that, so --"
"But what can we *do?*"
We went through a few rounds of this, with this poor kid just repeating the same question in different tones of voice, like an acting coach demonstrating the five stages of grieving using nothing but inflection.
It was an uncomfortable moment, and there was some decidedly nervous chuckling around the room as we pondered the coming AI (economic) apocalypse, and the fate of this kid graduating with mid-six-figure debts into an economy of ashes and rubble.
I firmly believe the (economic) AI apocalypse is coming. These companies are not profitable. They can't be profitable.
They keep the lights on by soaking up hundreds of billions of dollars in other people's money and then lighting it on fire. Eventually those other people are going to want to see a return on their investment, and when they don't get it, they will halt the flow of billions of dollars. Anything that can't go on forever eventually stops.
This isn't like the early days of the web, or Amazon, or any of those other big winners that lost money before becoming profitable. Those were all propositions with excellent "unit economics" - they got cheaper with every successive technological generation, and the more customers they added, the more profitable they became.
AI companies have - in the memorable phraseology of Ed Zitron - "dogshit unit-economics." Each generation of AI has been vastly more expensive than the previous one, and each new AI customer makes the AI companies lose *more* money:
This week, no less than the *Wall Street Journal* published a lengthy, well-reported story (by Eliot Brown and Robbie Whelan) on the catastrophic finances of AI companies:
The *WSJ* writers compare the AI bubble to other bubbles, like Worldcom's fraud-soaked fiber optic bonanza (which saw the company's CEO sent to prison, where he eventually died), and conclude that the AI bubble is *vastly* larger than any other bubble in recent history.
The data-center buildout has genuinely absurd finances - there are data-center companies that are collateralizing their loans by staking their giant Nvidia GPUs as collateral.
This is wild: there's pretty much nothing (apart from fresh-caught fish) that loses its value faster than silicon chips. That goes *triple* for GPUs used in AI data-centers, where it's normal for tens of thousands of chips to burn out over a single, 54-day training run:
That barely scratches the surface of the funny accounting in the AI bubble. Microsoft "invests" in Openai by giving the company free access to its servers. Openai reports this as a ten billion dollar investment, then redeems these "tokens" at Microsoft's data-centers. Microsoft then books this as ten billion in revenue.
That's par for the course in AI, where it's normal for Nvidia to "invest" tens of billions in a data-center company, which then spends that investment buying Nvidia chips. The the same chunk of money being energetically passed back and forth between these closely related companies, all of which claim it as investment, as an asset, or as revenue (or all three).
The *Journal* quotes David Cahn, a VC from Sequoia, who says that for AI companies to become profitable, they would have to sell us *$800 billion* worth of services *over the life of today's data centers and GPUs*. Not only is that a very large number - it's also a very short time. AI bosses themselves will tell you that these data centers and GPUs will be obsolete practically from the moment they start operating.
Bain & Co says that the only way to make today's AI investments profitable is for the sector to bring in *$2 trillion* by 2030 (the *Journal* notes that this is more than the combined revenue of Amazon, Google, Microsoft, Apple Nvidia and Meta):
How much money is the AI industry making? Morgan Stanley says it's $45b/year. But that $45b is based on the AI industry's own exceedingly cooked books, where *annual* revenue is actually *annualized* revenue, an accounting scam whereby a company chooses its best single revenue month and multiplies it by 12, even if that month is a wild outlier:
Hey! Before we go any further — if you want to support my work, please sign up for the premium version of Where’s Your Ed At, it’s a $7-a-month (or $70-a-year) paid product where every week you get a premium newsletter, all while supporting my free w…
Industry darlings like Coreweave (a middleman that rents out data-centers) are sitting on *massive* piles of debt, secured by short-term deals with tech companies that run out *long* before the debts can be repaid. If they can't find a bunch of new clients in a couple short years, they will default and collapse.
The AI bubble has absorbed more of the country's wealth and represents more of its economy than historic nation-shattering bubbles, like the 19th century UK rail bubble. A much-discussed MIT paper found that 95% of companies that had tried AI had either nothing to show for it, or experienced a loss:
Anything that can't go on forever eventually stops. Trump might bail out the AI companies, but for how long? They are incinerating money faster than practically any other human endeavor in history, with precious little to show for it.
During my stay at Cornell, one of the people responsible for the university's AI strategy asked me what I thought the university should be doing about AI. I told them that they should be planning to absorb the productive residue that will be left behind after the bubble bursts:
Of course AI is a bubble. It has all the hallmarks of a classic tech bubble. Pick up a rental car at SFO and drive in either direction on the 101 – north to San Francisco, south to Palo Alto – and …
Plan for a future where you can buy GPUs for ten cents on the dollar, where there's a buyer's market for hiring skilled applied statisticians, and where there's a ton of extremely promising open source models that have barely been optimized and have *vast* potential for improvement.
There's plenty of useful things you can do with AI. But AI is (as Princeton's Arvind Narayanan and Sayash Kapoor, authors of *AI Snake Oil* put it), a *normal* technology:
That doesn't mean "nothing to see here, move on." It means that AI isn't the bow-wave of "impending superintelligence." Nor is it going to deliver "humanlike intelligence."
It's a grab-bag of useful (sometimes very useful) tools that can sometimes make workers' lives better, when workers get to decide how and when they're used.
The most important thing about AI isn't its technical capabilities or limitations. The most important thing is the investor story and the ensuing mania that has teed up an economical catastrophe that will harm hundreds of millions or even billions of people. AI isn't going to wake up, become superintelligent and turn you into paperclips - but rich people with AI investor psychosis are almost certainly going to make you much, much poorer.
What if we repurposed those data centers to something more useful? What if the workers in those companies organized and unionized and stiked? Wouldn't that work?
@narF I think it would be hard to repurpose those datacenters. They're really geared towards running LLMs and that's mostly just GPUs and RAM to run the models in memory... a far cry from the "generalist" machines we usually see available in the "public" cloud that are actually useful.
@scarpentier @narF And from what I've read on the subject, it's not even *good* GPUs: LLMs mostly use FP16, so their hardware has crippled/limited FP64 (aka the really good stuff).
thanks for this. It reminded me of a proverb: -A fool with a tool is still a fool.-
The new economy bubble burst as well and lots of companies from then are gone. But still something is left. And not a little bit. I guess this will be the case here as well. On the other hand, Google, apple, Meta have loads of cash. Companies like anthropic or openai or or or may be soon sucked up, maybe by MS, maybe by Apple etc.
What worries me most are two things: 1. We use AI and vast resources to create artificial cat videos, summarize phishing emails and many other things not adding real value to our lives. 2. I believe I read and heard, from smarter people, that the transformer model architecture is not sufficient. But we stopped researching for new solutions.
China's still investing in core science and technology, betting on coming out on the other side a winner. I'm sick of the SV innovation hype bullshit, perhaps a hard reset will do it some good.
I work in the front line trenches of this bubble and I think what you say is correct. I do think that this is the result of the need to burn off excess capital, shrink the market, and then give the system room to grow again.
We are in for a collapse that will hurt most of us but allow the biggest players to consolidate further. Tens of people will make out like bandits (and not in a good way) and tens of millions will suffer.
After all of it, there will be some really useful tools that can be used to make really useful things. But they will all be owned by a few.
I hope I'm wrong and AI just collapses back to what it used to be - fun and interesting.
Quella in passerella oggi a Milano "è una collezione che celebra la vita, la vita naturale, la vita vera" sottolinea Nicoletta Spagnoli, anima del brand Luisa Spagnoli. (ANSA)
Uno stile leggero e leggiadro come la farfalla futurista dipinta dal maestro Giacomo Balla è protagonista della nuova collezione Laura Biagiotti per la Primavera/Estate 2026, che sfila nel Piccolo Teatro Grassi a Milano. (ANSA)
Il diavolo non veste più Prada, ma Dolce e Gabbana: la sfilata milanese del duo ha ospitato infatti le riprese del sequel dell'iconico film con al centro il mondo della moda. (ANSA)
Sagra dell’Uva di Mendrisio: 81 anni di tradizione Al via lo storico evento che dal 1944 unisce generazioni e celebra l’identità vinicola del Mendrisiotto con festa, musica e sapori locali rsi.ch/s/2259379
Migliaia di persone in piazza per la Palestina a Torino. Il corteo, diviso in diversi, folti, gruppi si dirige verso l'aeroporto di Caselle, nelle cui vinanze si trova anche uno stabilimento della...
The bully has filed a frivolous lawsuit against the New York Times demanding 16 billion dollars damages for publishing things he did not like. I say "frivolous" because lawyers say he has no valid grounds and can't win in a fair trial. The point, however, is to ensure it can't be fair.
MAGA GOP just hate vibrant, thriving communities. I recall that #Portland Oregon has great food choices, easy access to outdoor adventures, creative artists, clean streets, a variety of entertainment options, cool people, low crime rates, and ... No Sales Tax! Methinks Pedo POTUS is jealous. RE: bsky.app/profile/did:plc:lse7s…
Trump will die Luftbasis Bagram zurück tichyseinblick.de/kolumnen/aus… Vor kurzem erklärte der US-Präsident Trump überraschend, dass Washington den 2021 aufgegebenen Luftwaffenstützpunkt Bagram in Afghanistan wieder zurückhaben wolle. Zur Begründung verwies er auf dessen Nähe zu China. Der Stützpunkt liege „nur eine Stunde entfernt von den Orten, an denen China seine Atomwaffen baut“. Der ursprünglich von den Sowjets erbaute
In this exclusive interview, Pulitzer Prize–winning journalist Chris Hedges examines the authoritarian turn in U.S. politics in light of Charlie Kirk's assassination.
Report: The Hague Group convened 34 states at the UNGA to plan action against Israel:
Prevent military & dual-use exports Refuse Israeli weapons shipments at ports Prevent vessels carrying weapons to Israel Prevent support for Israel’s illegal occupation Pursue justice & hold perpetrators accountable Halt military procurement from Israel Divest public institutions from complicit companies Institute an energy embargo
As Israeli Prime Minister Benjamin Netanyahu took the stage at the United Nations General Assembly in New York on Friday, 34 states convened under the banner of The Hague Group to coordinate legal, diplomatic and economic measures aimed at halting Is…
Sarà un sacerdote della diocesi di San Miniato (Pisa), don Marco Billeri, 41enne di Pontedera, attualmente viceparroco nella parrocchia dei Santi Martino e Stefano a S. Miniato Basso, il secondo segretario personale di papa Leone XIV. (ANSA)
Profitez des vidéos et de la musique que vous aimez, mettez en ligne des contenus originaux, et partagez-les avec vos amis, vos proches et le monde entier.
Je construis des mots sur l'écran de pierre, Elle m'offre mille voix dans un silence lourd. Je grave mes pensées comme on taille le bois, Un lieu où la solitude règne ce jour.
Mes doigts frappent le clavier sur l'acier, Je forge des liens qui se brisent aussitôt. Elle montre des visages sans les toucher : n mur de connexions, prison de pixels.
Je parle dans le vide, sculpteur de vent, Elle répond par des cœurs, cicatrices.
Buongiorno, vi segnalo questo prossimo appuntamento di sostenibilità condivisa all'interno del programma della Cooperativa MAG2, per un invito e una richiesta di diffusione presso le vostre reti: mag2.it/notizie/754-condivider…
Un incontro con consigli pratici per creare una Comunità energetica rinnovabile e solidale (CERS) con l’obiettivo di ridurre le emissioni di CO2 e di inquinanti, risparmiare sulla bolletta, aumentare la socialità e aiutare le persone più fragili
Un…
This sculpture was created by Kelecsenyi Gergely & engineer Jozsef Szita. The fountain uses water to create the illusion of turning pages, and serves as a reminder of the importance of physical books.
Location: Eötvös Loránd University in Budapest, Hungary. On Egyetem Square, in front of the university.
„At the request of Secretary of Homeland Security, Kristi Noem, I am directing Secretary of #War, Pete Hegseth, to provide all necessary Troops to protect War ravaged #Portland, and any of our ICE Facilities under siege from attack by #Antifa, and other domestic terrorists. I am also authorizing Full Force, if necessary.“
Sabato 27 settembre Brescia per la Palestina, corteo di massa con il Coordinamento Palestina di Brescia, da piazza Rovetta fino a via Lunga, sede bresciana dell’azienda armiera Leonardo, legata a doppio filo al genocidio e all’occupazione israeliana …
Ukrainian President Volodymyr Zelenskyy has announced that the Ukrainian delegation will hold technical talks in the United States between late September and early October on acquiring weapons.
La guerra per il futuro della ricerca è iniziata ⚔️ Perplexity lancia le sue API e sfida apertamente Google con un motore AI che offre risposte dirette, non link 🎯 🤯 #PerplexityAPI #GuerraAGoogle #AI
Perplexity Search API, un motore di ricerca AI che sfida il modello pubblicitario di Google offrendo risposte dirette. Analisi di una battaglia che potrebbe cambiare il futuro della ricerca online
Enjoyed the mini exhibition, Rearview Mirror: Liverpool–London–Paris, at the Gagosian gallery in London, a series of photos and contact sheets taken by Paul McCartney just before The Beatles went global. Catch it before it closes.
Cory @pluralistic Doctorow in a must-read on Apple, the EU, privacy and the absolute need to get rid of facist-adjacent (or fully fascist) tech infrastructure.
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
A week ago, I turned that book into a speech, which I delivered as the annual Nordlander Memorial Lecture at Cornell, where I'm an AD White Professor-at-Large. This was my first-ever speech about AI and I wasn't sure how it would go over, but thankfully, it went great and sparked a lively Q&A.
2/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
One of those questions came from a young man who said something like "So, you're saying a third of the stock market is tied up in seven AI companies that have no way to become profitable and that this is a bubble that's going to burst and take the whole economy with it?"
I said, "Yes, that's right."
He said, "OK, but what can we do about that?"
3/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
So I re-iterated the book's thesis: that the AI bubble is driven by monopolists who've conquered their markets and have no more growth potential, who are desperate to convince investors that they can continue to grow by moving into some other sector, e.g. "pivot to video," crypto, blockchain, NFTs, AI, and now "super-intelligence."
4/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Further: the topline growth that AI companies are selling comes from replacing most workers with AI, and re-tasking the surviving workers as AI babysitters ("humans in the loop"), which won't work. Finally: AI *cannot* do your job, but an AI salesman can *100%* convince your boss to fire you and replace you with an AI that *can't* do your job.
5/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
When the bubble bursts, the money-hemorrhaging "foundation models" will be shut off and we'll lose the AI that can't do your job, *and* you will be long gone, retrained or retired or "discouraged" and out of the labor market, and *no one* will do your job. AI is the asbestos we are shoveling into the walls of our society and our descendants will be digging it out for generations:
pluralistic.net/2025/05/27/ran…
6/
Pluralistic: AI turns Amazon coders into Amazon warehouse workers (27 May 2025) – Pluralistic: Daily links from Cory Doctorow
pluralistic.netCory Doctorow
in reply to Cory Doctorow • • •Sensitive content
The only thing (I said) that we can do about this is to puncture the AI bubble *as soon as possible*, to halt this before it progresses any further and to head off the accumulation of social and economic debt. To do that, we have to take aim at the *material basis* for the AI bubble (creating a growth story by claiming that defective AI can do your job).
"OK," the young man said, "but what can we *do* about the crash?" He was clearly very worried.
7/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
"I don't think there's anything we can do about that. I think it's already locked in. I mean, maybe if we had a different government, they'd fund a jobs guarantee to pull us out of it, but I don't think Trump'll do that, so --"
"But what can we *do?*"
We went through a few rounds of this, with this poor kid just repeating the same question in different tones of voice, like an acting coach demonstrating the five stages of grieving using nothing but inflection.
8/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
It was an uncomfortable moment, and there was some decidedly nervous chuckling around the room as we pondered the coming AI (economic) apocalypse, and the fate of this kid graduating with mid-six-figure debts into an economy of ashes and rubble.
I firmly believe the (economic) AI apocalypse is coming. These companies are not profitable. They can't be profitable.
9/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
They keep the lights on by soaking up hundreds of billions of dollars in other people's money and then lighting it on fire. Eventually those other people are going to want to see a return on their investment, and when they don't get it, they will halt the flow of billions of dollars. Anything that can't go on forever eventually stops.
10/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
This isn't like the early days of the web, or Amazon, or any of those other big winners that lost money before becoming profitable. Those were all propositions with excellent "unit economics" - they got cheaper with every successive technological generation, and the more customers they added, the more profitable they became.
11/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
AI companies have - in the memorable phraseology of Ed Zitron - "dogshit unit-economics." Each generation of AI has been vastly more expensive than the previous one, and each new AI customer makes the AI companies lose *more* money:
pluralistic.net/2025/06/30/acc…
This week, no less than the *Wall Street Journal* published a lengthy, well-reported story (by Eliot Brown and Robbie Whelan) on the catastrophic finances of AI companies:
wsj.com/tech/ai/ai-bubble-buil…
12/
Pluralistic: How much (little) are the AI companies making? (30 Jun 2025) – Pluralistic: Daily links from Cory Doctorow
pluralistic.netCory Doctorow
in reply to Cory Doctorow • • •Sensitive content
The *WSJ* writers compare the AI bubble to other bubbles, like Worldcom's fraud-soaked fiber optic bonanza (which saw the company's CEO sent to prison, where he eventually died), and conclude that the AI bubble is *vastly* larger than any other bubble in recent history.
The data-center buildout has genuinely absurd finances - there are data-center companies that are collateralizing their loans by staking their giant Nvidia GPUs as collateral.
13/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
This is wild: there's pretty much nothing (apart from fresh-caught fish) that loses its value faster than silicon chips. That goes *triple* for GPUs used in AI data-centers, where it's normal for tens of thousands of chips to burn out over a single, 54-day training run:
techblog.comsoc.org/2024/11/25…
Talk about sweating your assets!
14/
Superclusters of Nvidia GPU/AI chips combined with end-to-end network platforms to create next generation data centers – IEEE ComSoc Technology Blog
techblog.comsoc.orgCory Doctorow
in reply to Cory Doctorow • • •Sensitive content
That barely scratches the surface of the funny accounting in the AI bubble. Microsoft "invests" in Openai by giving the company free access to its servers. Openai reports this as a ten billion dollar investment, then redeems these "tokens" at Microsoft's data-centers. Microsoft then books this as ten billion in revenue.
15/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
That's par for the course in AI, where it's normal for Nvidia to "invest" tens of billions in a data-center company, which then spends that investment buying Nvidia chips. The the same chunk of money being energetically passed back and forth between these closely related companies, all of which claim it as investment, as an asset, or as revenue (or all three).
16/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
The *Journal* quotes David Cahn, a VC from Sequoia, who says that for AI companies to become profitable, they would have to sell us *$800 billion* worth of services *over the life of today's data centers and GPUs*. Not only is that a very large number - it's also a very short time. AI bosses themselves will tell you that these data centers and GPUs will be obsolete practically from the moment they start operating.
17/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Mark Zuckerberg says he's prepared to waste "a couple hundred billion dollars" on misspent AI investments:
businessinsider.com/mark-zucke…
Bain & Co says that the only way to make today's AI investments profitable is for the sector to bring in *$2 trillion* by 2030 (the *Journal* notes that this is more than the combined revenue of Amazon, Google, Microsoft, Apple Nvidia and Meta):
https://www.bain.com/about/media-center/press-releases/20252/$2-trillion-in-new-revenue-needed-to-fund-ais-scaling-trend---bain--companys-6th-annual-global-technology-report/
18/
Why Zuckerberg says risking billions on AI is worth it
Lee Chong Ming (Business Insider)Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
How much money is the AI industry making? Morgan Stanley says it's $45b/year. But that $45b is based on the AI industry's own exceedingly cooked books, where *annual* revenue is actually *annualized* revenue, an accounting scam whereby a company chooses its best single revenue month and multiplies it by 12, even if that month is a wild outlier:
wheresyoured.at/the-haters-gui…
19/
The Hater's Guide To The AI Bubble
Edward Zitron (Ed Zitron's Where's Your Ed At)Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Industry darlings like Coreweave (a middleman that rents out data-centers) are sitting on *massive* piles of debt, secured by short-term deals with tech companies that run out *long* before the debts can be repaid. If they can't find a bunch of new clients in a couple short years, they will default and collapse.
20/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
The AI bubble has absorbed more of the country's wealth and represents more of its economy than historic nation-shattering bubbles, like the 19th century UK rail bubble. A much-discussed MIT paper found that 95% of companies that had tried AI had either nothing to show for it, or experienced a loss:
technologyreview.com/2019/01/2…
A less well-known U Chicago paper finds that AI has "no significant impact on workers’ earnings, recorded hours, or wages":
papers.ssrn.com/sol3/papers.cf…
21/
We analyzed 16,625 papers to figure out where AI is headed next
Karen Hao (MIT Technology Review)Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Anything that can't go on forever eventually stops. Trump might bail out the AI companies, but for how long? They are incinerating money faster than practically any other human endeavor in history, with precious little to show for it.
22/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
During my stay at Cornell, one of the people responsible for the university's AI strategy asked me what I thought the university should be doing about AI. I told them that they should be planning to absorb the productive residue that will be left behind after the bubble bursts:
locusmag.com/feature/commentar…
23/
Cory Doctorow: What Kind of Bubble is AI?
Locus OnlineCory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Plan for a future where you can buy GPUs for ten cents on the dollar, where there's a buyer's market for hiring skilled applied statisticians, and where there's a ton of extremely promising open source models that have barely been optimized and have *vast* potential for improvement.
There's plenty of useful things you can do with AI. But AI is (as Princeton's Arvind Narayanan and Sayash Kapoor, authors of *AI Snake Oil* put it), a *normal* technology:
knightcolumbia.org/content/ai-…
24/
AI as Normal Technology
Arvind Narayanan & Sayash Kapoor (Knight First Amendment Institute)Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
That doesn't mean "nothing to see here, move on." It means that AI isn't the bow-wave of "impending superintelligence." Nor is it going to deliver "humanlike intelligence."
It's a grab-bag of useful (sometimes very useful) tools that can sometimes make workers' lives better, when workers get to decide how and when they're used.
25/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
The most important thing about AI isn't its technical capabilities or limitations. The most important thing is the investor story and the ensuing mania that has teed up an economical catastrophe that will harm hundreds of millions or even billions of people. AI isn't going to wake up, become superintelligent and turn you into paperclips - but rich people with AI investor psychosis are almost certainly going to make you much, much poorer.
26/
Cory Doctorow
in reply to Cory Doctorow • • •Sensitive content
Image:
TechCrunch
commons.wikimedia.org/wiki/Fil…
CC BY 2.0
creativecommons.org/licenses/b…
--
Cryteria (modified)
commons.wikimedia.org/wiki/Fil…
CC BY 3.0
creativecommons.org/licenses/b…
eof/
File:Sam Altman - TechCrunch Disrupt SF 2017 (36522988343).jpg - Wikimedia Commons
commons.wikimedia.orghuntingdon
in reply to Cory Doctorow • • •Sensitive content
Cory Doctorow reshared this.
jokeyrhyme
in reply to Cory Doctorow • • •Sensitive content
> Anything that can't go on forever eventually stops.
i wish this were true, but it seems we haven't yet seen the limit on how much money investors will pump into Uber
are LLM investors differently-smart or differently-patient compared to Uber investors? 🤷
narF 😵✌️
in reply to Cory Doctorow • • •Simon Carpentier
in reply to narF 😵✌️ • • •Emma-Martin 🇨🇦🤝🇵🇸🇺🇦
in reply to Simon Carpentier • • •@scarpentier @narF And from what I've read on the subject, it's not even *good* GPUs: LLMs mostly use FP16, so their hardware has crippled/limited FP64 (aka the really good stuff).
Basically e-waste when the bubble pops.
narF 😵✌️
in reply to Emma-Martin 🇨🇦🤝🇵🇸🇺🇦 • • •unlofl [Promoted Toot]
in reply to narF 😵✌️ • • •Pteryx the Puzzle Secretary
in reply to Cory Doctorow • • •Sensitive content
A_Minion
in reply to Cory Doctorow • • •Sensitive content
Cory Doctorow reshared this.
Captain Superlative
in reply to Cory Doctorow • • •Simon B
in reply to Cory Doctorow • • •Okuna
in reply to Cory Doctorow • • •thanks for this.
It reminded me of a proverb:
-A fool with a tool is still a fool.-
The new economy bubble burst as well and lots of companies from then are gone. But still something is left. And not a little bit.
I guess this will be the case here as well.
On the other hand, Google, apple, Meta have loads of cash. Companies like anthropic or openai or or or may be soon sucked up, maybe by MS, maybe by Apple etc.
What worries me most are two things:
1. We use AI and vast resources to create artificial cat videos, summarize phishing emails and many other things not adding real value to our lives.
2. I believe I read and heard, from smarter people, that the transformer model architecture is not sufficient. But we stopped researching for new solutions.
Cory Doctorow reshared this.
A_Minion
in reply to Cory Doctorow • • •Allan
in reply to Cory Doctorow • • •Cory Doctorow reshared this.
deepfryed
in reply to Cory Doctorow • • •Cory Doctorow reshared this.
Flipper 🐬
in reply to Cory Doctorow • • •I work in the front line trenches of this bubble and I think what you say is correct. I do think that this is the result of the need to burn off excess capital, shrink the market, and then give the system room to grow again.
We are in for a collapse that will hurt most of us but allow the biggest players to consolidate further. Tens of people will make out like bandits (and not in a good way) and tens of millions will suffer.
After all of it, there will be some really useful tools that can be used to make really useful things. But they will all be owned by a few.
I hope I'm wrong and AI just collapses back to what it used to be - fun and interesting.
Janik
in reply to Cory Doctorow • • •> AI is the asbestos we are shoveling into the walls of our society and our descendants will be digging it out for generations.
Absolutely!