Ever since Greenspan took over the Fed and the 87 crash when they figured out their playbook, the US has only had unavoidable stock market crashes. The Fed is always there to juice markets higher and to jump in at the least sign of a normal (pre-Greenspan) market correction.
But sometimes the irrational stupidity overwhelms even the Fed, because they are both stupid and ideologically unwilling to ever force a correction. This happened twice: the dot-com boom and crash and the Mortgage backed security boom and crash (if we bundle shitty mortgages based on lies together, they become not shitty, because we’re pretending they aren’t all basically the same thing!)
Now we’re going to get the AI Boom crash. I’m well over 90% on this. The AI booms is in the “wildly stupid over-claiming” stage. It’s not that token based AI isn’t a real tech, or that it doesn’t have some uses, but the claims of it completely changing everything (replacing a third of the workforce, acting without human help to run things, being able to cure cancer and make huge theoretical breakthroughs) are obvious over-reaches. So far every academic study that comes in shows that AI isn’t even good at the one thing everyone anecdotally agreed it was good at: writing code. Right now it seems to mainly be a good way to cheat at university, to have a fake relationship, or to bypass Google’s shitty search (which is what I use it for.) It hallucinates, the hallucinations cannot be removed because they are integral to the tech, and the code it produces, even when it works, is a huge mess that will cause massive maintenance issues.
In addition:
- Since it doesn’t actually mostly reason, it requires data sets bigger than all the data in the world if it is to keep improving;
- If it uses the data it itself produces, it experiences model collapse.
- None of the American AI companies make money per query. Every query costs more than they can charge.
- It requires a vast build-out of energy and data centers, of the “over a trillion dollars” variety. There literally isn’t enough money to pay for OpenAI and Anthropic’s dreams, and there isn’t a product at the end of it that could pay back all that money.
- About 40% of the US stock market is now based around NVidia and the AI companies.
- NVidia has now invested in Open AI, so that they can turn around and buy more NVidia cards.
- The Chinese offer an open source AI which is almost as good and with costs somewhere between one fifteenth and one-thirtieth as much, so that it might actually be profitable AND since it’s open source, Trump can’t have a mini-stroke and decide to cut you off at his whim.
It’s my annual fundraiser. This allows us to cover the changeover of hegemony from America to China, environmental collapse, internal US fascism, what a better society would look like, Gaza, AI, the coming stock market crash and various other issues. As of this writing we’ve raised about $2,700 out of a $12,500 goal, from over 25 people. It’d be great if you can help out (please don’t donate if your financial situation is dire.) You can Subscribe or Donate here or contact me at admin-at-ianwelsh-dot-net if you need another way to donate (mail, usually. A lot of cash apps don’t work in Canada.)
Throwing all this money at AI if it really was the epochal “tech to end all techs, the singularity, dude” that the tech-bros claim it is might make sense. But I don’t see the evidence that this is the case, and even if it is, why not use the Open Source Chinese variety?
In fact, my guess is that this version of AI, based on this model and this generation of chips, is not even as big a deal as the internet was. Everyone was right that the internet was going to be HUGE, they just over-invested before it was and before people knew who the winners (Google, Facebook, Amazon) were going to be.
But so far AI doesn’t even look as important as the internet, but the spend is way larger than the internet build-out of the turn of the millennium.
But even if AI turns out to be a HUGE deal, it’s going to crash out of this bubble and we’ll find out later who can make money doing what.
The Fed will paper the AI market crash over, making hundreds of billions or even a trillion out of thin air to save the rich from their own stupidity and greed. Again. But this will be the LAST crash the Fed will be able to save the capitalists from. The one after will either wipe the capitalists out, wipe out America, or both.
Feral Finster
“The Fed will paper the AI market crash over, making hundreds of billions or even a trillion out of thin air to save the rich from their own stupidity and greed. Again.”
This much is obvious.
“But this will be the LAST crash the Fed will be able to save the capitalists from. The one after will either wipe the capitalists out, wipe out America, or both.”
We hear that before every big crash. What makes you so sure?
bruce wilder
I am glad you laid down a marker on this. It is so obviously coming.
Like you, I called the GFC and did so in late 2005, early 2006, when I was commenting on Mark Thoma’s blog almost every day. And, reading Calculated Risk every day — it didn’t require any personal genius, just being well-informed, which CR made me. It was like watching a train wreck. (The thing that makes a train wreck so surreal to watch is that you know no one can swerve.)
The weird part to me in retrospect was the pervasive woocoodenode, afterward. The powers-that-be brought in Bernanke rather forcibly and in a big rush, which was a big tell that someone with real power knew what was coming.
This time, it is much harder to read — for me at least. Partly because I am not so well-informed about a much more complex and unprecedented situation. I have been studying my basic economics and finance, but all it accomplishes is to confirm that a fundamental regime change is underway. And, to make me aware of strange subcultures of monetary reform living among the ex-hippies.
The AI bubble is ridiculous froth of course. What has the Powers-That-Be frightened is China, deflation and the collapse of dollar hegemony. Of course.
What scares me is that the Not-so-bright are aware, too, and believe crypto, Central Bank Digital Currency, RealID add up to a solution. Maybe war with Russia thrown into the mix. Or counter-revolution against open-borders immigration?
The last time, in 2005-9, Job #1 was to out flank righteous indignation about pervasive fraud so the ill-gotten gains could stay gotten and no reform that meant anything could be instituted. A simple story really.
Now, there isn’t even mind-space for a righteous reform, as far as I can see. There is scarcely a memory in the U.S. that a left ever existed and all the populist energy is leaking out on the right. All I can think is that line from GOT: “chaos is a ladder”.
Forecasting Intelligence
Agree entirely Ian. Great post.
Jefferson Hamilton
“But this will be the LAST crash the Fed will be able to save the capitalists from.”
Why?
different clue
I just hope AI crashes all the way to ground level before the transhumanists ( and other secret-agenda AI pushers) are able to push the earth beyond all the climate tipping points of no return.
It is worth exterminating the economy to save the ecology, if that is the choice we are forced to endure.
WangoTango
Never underestimate the power of the bailouts. They are all there is left.
different clue
I too will claim that I had an idea that something really bad was going to happen before the GFC of 2008. I can’t prove that of course.
But here is my claim and my reason. I used to watch CSPAN when I still had cable and cable still showed CSPAN, before the Great Degradation of Comcast Cable. I saw Bernanke on there talking to the Congressmen, and right when he was about to explain why things were stable and nothing bad was happening or would happen, his voice cracked and almost broke from barely contained abject fear. Since a brilliant intellectual with big degrees like Bernanke thinks he is smarter than any 500-or-so Congressmen gathered in a room, I knew he wasn’t afraid of the Congressmen. So would could make him almost panic for a second or two before he collected himself?
He must be seeing something very big and very bad over the far horizon and coming our way. I had no idea what he was seeing or what it would be, but I could tell that he saw it, and he almost went into lose-it freakout before he recovered his poise.
That’s my claim that I knew “something bad” was going to happen . . . because I saw/heard that fractional second of terror emanating from Bernanke.
Mark Level
I believe you are correct. AI is the new Tulip Mania Bubble, has been for some time. Decades since I read about the Tulip thing, but evidently they became a speculative item and at one time a single tulip (accd. to what I read) could cost as much as a good sized house. Pump and dump, go USA baby!!
And here’s a tangential thing I want to vent on. How much can Trump spit in the face of his followers and still have any popularity. Jeffrey St. Clair recently did a long list showing Trump is deeply unpopular in a majority of Red states. He is now spitting in the CHUDs faces, openly. A couple obvious examples– 1. In 2016 he ran to the Left of $Hrillary on foreign policy (Not hard to do, “We came, we saw, he DIED! Wahahooo haha!”) and demolished Jeb and the Bush Dynasty. Jeb later whined, convincingly, “I had no political viability”, his moron brother destroyed it. Tens of millions in corporate donations and he got 1 delegate at the RNC!! So NOW Trump is best buds with Bush Jr’s shoe-cleaner (with his lips) Tony Blair, aka Tony B.liar to the Brits who hate him, to co-head his Gaza “Peace” (sic: Final Solution) Committee. The Rethuglican right hated Bush and DJT pretended to be “different”. Nope, a fatter carbon copy.
2. I just saw in my Firefox “pocket” that Trump is teaming up with and promoting Pfizer!!! https://www.npr.org/sections/shots-health-news/2025/09/30/nx-s1-5558432/drug-prices-trumprx-pfizer?utm_source=firefox-newtab-en-us This is so obviously a criminal scam, it’s not even hidden. Ironically NPR reports it, but I doubt it’ll make their Pod People admire Trump for promoting Pfizer. Why have the heroin addict/ clown RFK Jr. as your MAHA expert being denounced by the Libs and meanwhile you cut a criminal deal with Pfizer.
Biden sure betrayed and shat on the few actual “progressives” in the Dimmie party, didn’t matter. Will it matter in the case of the Red Meat R’s– I am guessing yes. He’s about to burn all this shit down evidently. At this point it looks like the late shock artist (he was a crappy musician so he got audiences via self-mutilation, scatology etc.) GG Allin is running things. Herman Cain died in vain!!
Ian Welsh
If anyone cares the proof that I predicted the financial collapse is in the wayback machine. Or Sean-Paul can attest and thousands of people read the article.
Feral: I didn’t say that before the last two crashes. Why I am (fairly) certain is that doing so requires either near autarchy OR currency privilege and when the next one happens the US will have neither. Of course they can print the money anyway, but if they do it will have horrific consequences.
Purple Library Guy
I’ll take a stab at “why”: I’m thinking that the main reason the US central bank can get away with just throwing unimaginable amounts of money at this kind of problem is because the demand for US dollars has traditionally been very high and global because of the whole reserve currency thing, and because of the “flight to safety” psychological thing, where in the past economic crises, EVEN ONES CAUSED BY THE US, have resulted in panic buying of US dollars. Both of these patterns are waning; I think even doing it this time is likely to have a lot more consequences than in the past.
By the time we get to the next more-or-less-separate crisis, the dollar will certainly have much less room to maneuver, and money-printing on that scale all at once in a public way will likely result in a major inflation shock as the usual buyers just . . . don’t.
There are other issues, though–the AI crisis is a weird one. I mean, it feels oddly detached from the real economy, and yet I get the impression that the AI bubble grew in the first place as a process of desperately papering over a bunch of other economic weaknesses. When AI goes down, it could cascade into a bunch of other failures, and while a big bailout might sort of fix the AI thing by itself, who knows if it will be enough to deal with the other ones. Like, if the confidence fairy goes away, what will happen to crypto? Property, including commercial? Various kinds of debt, like student? How many USians are on the edge of not being able to pay medical insurance; if they stop, what does that do to overleveraged private-equity medical-stuff companies? There’s a bunch of stuff that could go wrong; if the music stops for AI we may discover there’s a lot more than one chair missing as we rush for the seats.
Feral Finster
“I didn’t say that before the last two crashes. Why I am (fairly) certain is that doing so requires either near autarchy OR currency privilege and when the next one happens the US will have neither. ”
I believe you regarding predicting the last two crashes. What i learned from those crashes (I predicted them as well) was that people of influence and authority worldwide (China included) do not want the current system to fail and they will do whatever it takes to prevent it from failing.
Us peons don’t count. We never did.
Ian Welsh
Back then (almost) everyone needed the US. They may still. They won’t next time.
different clue
@Feral Finster,
Do us peons count to ourselves? Do we count to eachother?
Feral Finster
@different clue:
“Do us peons count to ourselves? Do we count to eachother?”
Generally no, not that it matters.
Feral Finster
@Ian Welsh
I’d like to believe, but the US is a natural empire as a simple matter of geography
mago
Ineptitude, buffoonery, shills and marks. It’s like a carnival fairway. Step right up and take a look at the two headed lady. Bend right over and kiss your ass goodbye. AI has got you covered.
And if you doubt and disbelieve there’s a concentration camp cell for you on the dark side of the moon.
We’ve got your number.
different clue
@Feral Finster,
Well, that sounds like two things us peons need to change . . . mattering to ourselves and mattering to eachother.
How many peons would have to care about themselves and eachother in order to make a difference? Is it an experiment worth running?
Feral Finster
@Different clue
I think it was Jay Gould who dismissed the concerns of peons, saying that he could always hire one half to kill the other half.
Anyway, foreign holdings of US government debt hit an all time high recently.
Tailhedgetrader
Great post. What makes you think the next crash won’t be the last one? I feel like it will be the mother of all asset bubbles popping led by ai crash as a catylst. After that, we’ll be lucky to have 2% avg annual growth if that.
Like & Subscribe
When are the “radical leftists” going to start disabling all the data centers? Ah, that’s right, they’re all busy nurturing their social media presence and cache and plus or besides, they’ll let the AI crash take care of it so they don’t have to do the dirty work, oh yeah.
Flaser
RE: The AI Gospel
People who subscribe to this tend toward transhumanism AKA the Singularity.
Contrary to their expectations, such an event would likely be more akin to Cthulhu waking than the Rapture.
(Insert the collected works of Peter Watts to highlight why: *non*-human intelligence is unlikely to be banally evil like Skynet, instead it’d almost certainly be *amoral* by human standards).
Speaking of which: it is indeed the latter dressed up in a new “scientific” coat that this lot seeks. In other words: cosmism with a fresh coat of paint.
Charles Stoss has some choice words about this:
https://www.antipope.org/charlie/blog-static/2011/07/federovs-rapture.html
When viewed in this light AI optimism is stripped of a lot of its seeming scientific authority.
Gaianne
Ian–
Thank you for the clearest and most concise description of AI that I have yet encountered.
–Gaianne
GramSci
A quibble: humans ‘hallucinate’ way more than LLMs: pick 3 Trump tweets at random. Hell, pick 3 random ‘human’ tweets.
The cras: yeah, it’s been happening to the dollar. But will Jerome Powell do the Messiah act before the midterms?
The silver lining: as PlutoniumKun noted on NC, the additional generating capacity. Ten years from now, if we’re still graced with his presence, Donald Trump will claim he invented electrification.
steppenwolf fetchit
@Flaser,
I haven’t studied the transhumanists or the singularians ( like Raymond Kurzweil of The Singularity Is Near . . . https://en.wikipedia.org/wiki/The_Singularity_Is_Near . . . and now, The Singularity Is Nearer . . . https://en.wikipedia.org/wiki/The_Singularity_Is_Nearer ).
My understanding is that these people believe in creating a Super Duper Grand UltiMega AI so they can upload their entire brain-mind consciousness-knowledge-very soul-etc. into-onto the UltiMega AI data storage facilities. Then their true selves, set free from their meatly bodies, can live forever inside the beautiful machine.
Then they won’t need biological systems anymore because the beautiful AI machine will keep their uploaded minds alive without any biology at all. They believe that the earth and all its animal-mineral-vegetable creation-inventory exist to support building this UltiMega AI so they can upload themselves into it. They are happy to autoclave the planet earth itself if that can get their AI created and themselves uploaded into it.
Autoclaving the earth may well be their secondary goal all in itself, as their last laugh upon the rest of us.
The only way to stop them would be for society at large to round up and physically exterminate all the transhumanists and all the singularians and all the AI technologists and experts so that no one exists anymore upon the earth to seek AI or to build it.
And who among us would dare to suggest such a thing, let alone call for it?
So the best we can hope for is a giant AI bubble pop crash which puts the entire world, China included, into the Last Best Greatest Depression Ever of All Time. At least that would stop the Digital Lords of AI from turning Earth into Venus 2.0 . . . at least for a while.
mago
@steppenwolf fetchit. You’re overthinking it, just as the transhumanists are.
These people are intoxicated on their own hubris. Not going to bother defining and refuting their bullshit. That it’s taken as real is reification of delusion and illusions, which is SOP across the board for the most part.
Don’t buy the lies.
different clue
@Mago,
I didn’t say the singularian transhumanists would succeed at creating their super AI machine to load themselves up into. I just said that is what they want to do and are devoted to working for. It probably can’t be done and it probably won’t succeed.
But they can certainly succeed at burning so much more coal, gas and oil to fuel their computer farm attempts that they can certainly succeed at pushing earth irreversibly past the climate change tipping points. I don’t think I am overthinking that aspect of it at all.
So I still think a swift and soon AI bubbleburst and a Greatest Ever Depression would be a better thing than for the AI community to succeed in bringing back a Paleocene Eocene Thermal Maximum climate ( or even worse).
https://en.wikipedia.org/wiki/Paleocene%E2%80%93Eocene_Thermal_Maximum
Dan Kelly
I just had to post this somewhere:
‘banks do not drive foreign policy nor are they in the business of resource development’
– Yves Smith, October 3, 2025
https://www.nakedcapitalism.com/2025/10/trumps-imminent-wars-and-economic-damage-as-his-legitimacy-crumbles.html#comment-4293780
Yves’ entire statement:
‘I hate to tell you but what I have heard Alex Krainer say about the global financial crisis and other banking related matters is so often flat out false that I never listen to him.
And more specifically here, banks do not drive foreign policy nor are they in the business of resource development.
Even in Japan, where banks were sometimes the lead members of zaibatsu, later kairetsu (as in more powerful than Western banks), they did not drive business policies.’
—
At minimum, banks are an integral part of what drives business policies and development.
At minimum.
The following is a largely accurate summation from Wikipedia in which the word ‘bank’ appears over 30 times as being at minimum part and parcel – but often head of the hydra – of all the activity that went on in the Keiretsu economy of business and development:
https://en.wikipedia.org/wiki/Keiretsu
—
I don’t listen to Alex Krainer either.
Dan Kelly
I just had to post this somewhere:
‘banks do not drive foreign policy nor are they in the business of resource development’
– Yves Smith, October 3, 2025
https://www.nakedcapitalism.com/2025/10/trumps-imminent-wars-and-economic-damage-as-his-legitimacy-crumbles.html#comment-4293780
Yves’ entire statement:
‘I hate to tell you but what I have heard Alex Krainer say about the global financial crisis and other banking related matters is so often flat out false that I never listen to him.
And more specifically here, banks do not drive foreign policy nor are they in the business of resource development.
Even in Japan, where banks were sometimes the lead members of zaibatsu, later kairetsu (as in more powerful than Western banks), they did not drive business policies.’
—
At minimum, banks are an integral part of what drives business policies and development.
At minimum.
The following is a largely accurate summation from Wikipedia in which the word ‘bank’ appears over 30 times as being at minimum part and parcel – but often head of the hydra – of all the activity that went on in the Keiretsu economy of business and development:
https://en.wikipedia.org/wiki/Keiretsu
—
I don’t listen to Alex Krainer either.
bruce wilder
@ Dan Kelly
Sometimes, Yves Smith is just razor sharp in cutting through an argument, but sometimes she seems peculiarly prone to declarative assertions that could not possibly be true, from any conceivable realistic point of view. What went thru her mind when she was coming up with this formula of words. The most hackish AI bot would not hallucinate so artlessly and implausibly.
I looked for some evidence of ironic or sarcastic intent or an omitted negative (things I am prone to do, to frequent embarrassment). Nada.
Thanks for posting this, by the way. It is bizarre but common reality is so 20th century politics anyway . . .
Jorge
After the GFC, Fox News creeps came up with the racist, classist explanation that poor people were buying houses which they should not have (true) and that the Feds forced the banking system to fund them (wildly false). Currently we often hear that Covid-flation happened solely because the dirty poors were given The Free Money while unable to work- no greedflation, no supply chain shocks involved.
What racist, classist explanations will appear for the AI Crash?
p.s. “Did the Great Depression happen because those Heathen Celestials eat so much gold in their exotic restaurants? We’re just asking questions.”
Jorge
And separately- the various crashes from overbuilding infrastructure (railroads, dot-com fiber bandwidth) have been followed by new ventures which collected partly built infrastructure in low-debt businesses, finished the buildouts, and operated the resulting infrastructure. What wreckage will be left by the AI Crash, and how can it be repurposed in financially sound ways?
* AI model IP will be left behind (Mistral, OpenAI, HuggingFace are obvious candidates). These are somewhat like fresh vegetables; they don’t automatically update.
* Large data centers are really interesting. Suppose you have a giant, mostly-empty building in the middle of nowhere with robust power and water connections. Hydroponics? Cover the roof with solar? (After all, the builders have already destroyed the farmland.) Both? Since it will have robust security, growing pot is an obvious use.
* A giant data center competely wired up, formerly owned by a defunct company, can serve as an inference farm for awhile, as the gear slowly fails and depreciates out.
* The kids who took $100 million pay packages. What strange religions will they start?
Jorge
And while we’re re-litigating the GFC, I’ve heard an interesting theory.
What happened in 2005? Peak Cheap Saudi oil. This is pretty solidly established. What’s the largest international mountain of money? Oil money, in particular Saudi oil money. The oil industry had to react, slowly, even if it didn’t want to. Under this theory, during the early and mid-Oughts, silent remote money avalanches started. This instability slowly crept closer to financial centers, and knocked over mid-sized mountains of unstable money in the mortgage markets. There were many other small mountains of unstable money in other countries: misallocated government bonds in Europe, misallocated life insurance money, reinsurance companies reacting to Climate Change. The Gulf Wars happened. The very rich were starting to realize the world was not what they thought it was, etc.
After all, it was not the Wall Street crash, it happened around the world.
Jorge
“AI Boom crash. I’m well over 90% on this.”
I’m obviously with you on this. The crash will happen, I’m sure. The length, depth and complexity are open. For example, it’s an open question about whether there will be a “Minsky Moment”: that terrible time when all big players are overextended, have to sell some assets to make good their other promises, and discover that there are no buyers for the normal kinds of assets. Instead of selling what they want to sell, they can only sell a few kinds of things that others will buy: US Treasuries, gold, commodity futures, Picassos & Warhols, very few other kinds of asset.
And the US is doing a fine job of pushing Treasuries from the top of this list.
different clue
WallStreetBets has posted a video of a Three Stooges short with the Stooges labeled ” OpenAI, NVIDIA, and AMD. The poster titled it: ” OpenAI, NVIDIA, and AMD, The Tech Triangle That’s Driving the Market ”
https://www.reddit.com/r/wallstreetbets/comments/1nzx2hv/openai_nvidia_and_amd_the_tech_triangle_thats/