You have alighted onto an interesting topic: “bail out”. I believe a closely related word on the semiotic chain is “retreat”.
I don’t feel that our culture would wish to “retreat” from AI. I know quite a few managers and programmers who absolutely delight in the fruits of the industry.
No doubt the titans of the AI industry, who stand to lose their shirts, would also like a bailout. As would the workers under their leadership. Finally, if the governments of the world are taking “AI” technologies as seriously as I think they are, then I should expect a bailout in the event of bubble go boom is inevitable.
In which case, Buy Low; Sell High.
But money talks, I guess.
And they’ll get away with it too
ie for every loser there is a winner.
Obvious ones are those who can extract value from this companies now ( high salaries/bonus etc ), swapping worthless paper for real assets, Nvidia and their share holders etc. Investors who hold short positions etc.
And since we are talking about USD (specifically "of" USA), it also means globally a hell lot more of "all of us" are on the hook than we would have been involved (even remotely) had (or would) this endeavour ever ended up in some sort of general success.
I don't know whether it'd be "tails I win, heads you lose" or "I reap the profits alone, you reap the losses alone". Maybe the latter and it becomes magnitudes more interesting when it expands (or rather engulfs) beyond the boundaries of the great nation.
If I owe the bank $1,750,000,000 and can't pay, it's the bank's problem.
Everyone likes to repeat this tired cliche but skips all of the steps that would have to happen to get to that point.
Channelling the 1980s for off-balance sheet financing 101.
From an economic perspective there is zero difference between borrowing to buy an asset and entering into a non-cancellable long term (equivalent to its economic life) lease for the asset.
The first option causes an asset and a liability on the balance sheet, affecting debt ratios that appear in financing contracts and so on. The second does not appear on the balance sheet.
You pay every month, like it or not. You call it interest or you call it a lease payment. You need it off balance sheet for reasons, investment bankers will structure that to make it happen for a fee.
Technically, from an economic perspective, it's debt.
Isn’t the important difference that it doesn’t trigger bankruptcy on default? Economically it might not be that different but it has some significance legally because the courts have some fast tracks that trigger bankruptcies (IANAL but that’s my layman’s understanding).
If they “default” in this case it will lead to lawsuits that they will almost certainly lose but in the mean time they kick the can down the road hoping to recover on general economic headwinds like lower interest rates. IMO the risks are obviously correlated here but I’m betting short term incentives drove this mess.
See Apple's FY2025 10-K, the leases are in page 42 under "Lease-Related Assets and Liabilities", which shows:
Operating leases
- Other current liabilities: $1,579 million
- Other non-current liabilities: $10,911 million
Finance leases
- Other current liabilities: $538 million
- Other non-current liabilities: $692 million
Total lease liabilities: $13,720 million
https://s2.q4cdn.com/470004039/files/doc_financials/2025/ar/...
The way banks get involved is that they may be lending senior financing to the private credit funds but that means they have a ton of subordination.
I believe that some of the private credit is the banks' own private credit arm! It's absolutely mental: the bank can't lend to them so they have a private credit company/division that can skirt around the rules and still provide the capital!
Also, I bet that the Oracle datacenters aren't for Oracle LLC but Oracle Datacenter No 4 LLC so that if it all goes tits up, Oracle LLC walks away!
Remember the vendor financing model which got a lot of technology companies into trouble at the same time? It took some years, almost all those companies disappeared. Motorola, Nortel, Lucent ...
> By investing in the data center's operating company with a 20% stake and using the facility under a lease agreement, Meta secured computing resources but also increased its hidden debt.
> Meta has a contract guaranteeing investors' losses if the data center becomes unnecessary and the lease is terminated.
The way they are hiding these debts is by having a stake in a data center company. But if shit goes tits up they are contractually liable for 100% of the losses.
It's not the fund that lent the money to the SPV, it is Meta who is offering guarantees here.
Banks have not been loaning AI money for some time. They hit all their regulatory safeguard limits so they can't keep loaning. Half the money being invested in AI is private capital. There is still systemic risk, because private capital is a shadow banking system and you don't know who will be affected when they go kaput. Your utility company may [read: will] go bankrupt, but the money in your personal bank account is safe. Your retirement account, however...
Privatize revenues, socialize costs of business
You're not "competitive" if you have to pay for workers, ink or investments. Workers only get a wage because of communist populists who don't understand business, and (insert place - even the Philippines these days) has the most highest taxes and wages around the world.
I mean if AI falls short we might see a collapse in the price of colo but those investments would probably just be paid back over 10x the period.
It's a tough job market out there and it took 4 months to land job offers after being laid off.
I have a competitive offer from Oracle OCI with a team adjacent to this initiative and I am seriously considering it. How long do you guys think it will take to blow up (if it does)
To the experienced devs out there; would you take a 15% less offer from a medium sized company for job security?
too often pessimists are treated as serious, critical thinkers, while optimists are dismissed as naive or foolish.
remember this guy Michael Burry who saw the 2008 housing crash coming, bet everything against it, and made a fortune?
but then he spent the next 15 years trying to pull off the exact same trick, lost a ton of money fighting a massive bull market, and finally had to shut down his hedge fund because lightning just wouldn't strike twice.
how I deal with this: focus on what you can control. reading news in general seems to be difficult. what can I personally do about all that? very little tbh.
In a hot job market, no.
In the current job market, yes.
It's one of the main reasons I am sticking to my current employer. I could likely make more money by moving jobs right now ; I still get the odd recruiter reaching out here and there; but I fancy the security when I see dark clouds on the horizon.
And oh boy, those clouds are very dark. Foreboding indeed.
Softbank always has their finger on the hype pulse ...
I will go against the grain of everyone else - be a hype beast, every crash I have seen they make more and get better jobs on the flip side. Counter Oracle at a higher salary and ask about bonus guarantees for when this all goes to the moon (you have to really really appear to believe).
But a dig +trace archive.ph (which recurses all the way to the root locally) resolves it fine.
Is there some US-mandated DNS filtering I don't know about?
But on the other hand, these companies are essentially paying for the service of taking the debt off books (by paying the leasing premium to the SPV partners). I guess I'm wondering what they really gain from doing so, if again sophisticated investors can see through the games?
But of course it is obvious in this scale. However, credit ratings do not care, as they are driven by regulation, and regulators get their paycheck regardless.
Also in the joint venture like Blueowl/Meta for the $27B Hyperion data center in the case when things go wrong Meta is in theory bankrupt remote. So in theory it should not affect credit rating because when the bad debt is not served, Blueowl, not Meta, is in the hook for it. And Meta's investors should be protected for this event.
The actual lending is done by private credit institutions that have raised money, sometimes on the order of 10s of billions of $.
But the main reason 2028 would be a risky year is the systemic risks around Trump. If he makes even more efforts to put his thumb on the electoral scales than he already has, or if it is not clear there will be a free and fair election (from an external perspective it really isn’t), then confidence drains out of the wider system very quickly, and one of the earliest things to go in such a situation is speculative investment.
Even if there is a free and non-controversial election, the market might well see risks for the AI companies: would the Democrats, if they make it into power, be so fully on board with letting these companies operate in a low-regulation environment?
And if the bubble should look like bursting for its own reasons in early 2028, will the by then quite stressed federal government, late in the term, actually have the real authority to do anything about it? It’s not a popular government, or a very organised one, and whatever they do will need confidence, co-ordination and an unprecedented level of buy-in
That means losses. Big losses for some.
I assume that these off-the-books companies can quite literally be pinched off and the debt becomes the banks' problem, so the primary company, i.e. Meta, Oracle, can walk away but the banks will be left holding the bag.
We know what happened the last time the banks played their stupid games!
There isn't always a mechanism that will translate being-right about some fundamental fact into making-money from it.
P.S. It isn't trivially "solved" by someone setting up a gambling scheme either. Many people who are right will still lose because they didn't also guess when other people would start to agree with the conclusion. Plus there's the time-value of money for things which take longer to play out.
According to Wikipedia, the Manhattan project only cost $28 billion in 2024 dollars. That's a completely different order of magnitude compared to what we're talking about here. I imagine the size of the investment here would even be hard for the US government to swallow.
I don't actually think the current administration, or the people behind it, actually care about real world economics anymore.
Both are true.
When it comes to military level stuff, it seems to exist outside of normal financial logic. They just invent money for it.
As for China - I think their motivation is clear: if they could have pulled it off from the start, their business model would have been the same as MANGO's. Llamacpp opened up the floodgates and their best bet now is to use this and their resources to start pulling the rug from underneath MANGO. I'm not a fan of the "enemy of my enemy" philosophy - they can both be enemies but wile they are at each-other's throats (albeit not openly), I'll gladly sit back and grab some pop-corn.
Further, unlike during the dotcom crisis, most of this spending is not driven by debt. It's mostly funded by the large companies which are producing enormous revenue and profit to pay for this. It's an order of magnitude different.
The major question mark on these valuations is the revenue assumptions, which can be reasonably criticised. A bear case here is that revenue growth is not as aggressive as projected, and valuations steadily decline over time. I don't see a likely scenario where the entire sector collapses. There is no apparent cascade failure mechanism. Of course, these mechanisms aren't always immediately clear prior to crashes.
We must remember that all of the models coming out of China are presumed to be distilled frontier models. Meaning a) they will always be x days/weeks/months behind the frontier models, b) they will never be quite as good, c) inference will generally be constrained by compute capacity (especially as the frontier studios have an incentive to capitalise on their moat), meaning Chinese studios will always be at a disadvantage.
The real wildcard here is self-improvements. It looks like we're already in the singularity, meaning a large proportion of LLM development is already done by LLMs. The development cycle on these might be months now, but it will be weeks soon. Days within a year, then hours, minutes, seconds, and milliseconds. It's impossible to predict what this curve looks like.
They'll be knackered though, unlike a good Herman Miller chair. :-)
We might even be able to afford RAM again!
Or rather could have been, since the issue seems like it was transient. I'm now getting resolutions.
Very strange, I wish I'd recorded what was going on better while it was broken.
If they want the reward, they need to accept the risk when the bet doesn’t go their way. If they aren’t willing to accept that, they shouldn’t be making such big bets.
It won't be framed as bailing out the AI companies. It will be sold to the public as bailing out their pensions and investments.
It's all about the marketing...
Public anger is a very pre-smartphone thing. You don’t worry about those sort of things when you’ve hyperoptimized the circus and made the bread cheap enough
That's putting it mildly, IMO.
An AI bailout would be a poison pill that would electorally doom whichever party was behind it for decades.
if Google had to pay a $2b bill they can spend $20m fighting it.
If they have to pay a $2t bill they can spend $20b fighting it
You can buy a lot more justice with $20b than with $2000
They are aristocrats. They are worried about accumulating more power then their peers, about alliances to gain that power, and about projecting an image of infallibility.
Aristocracy has never been known for caring for economics, technology or practical matters. They just care about power. Ruining nations, building useless vanity projects and going to war is kind of their main thing.
Nominal GDP from https://en.wikipedia.org/wiki/Economic_statistics_of_the_Uni...
But I'll be holding out for the gently used one owner B300.
The circus is distracting from the price of bread, apparently.
Public anger has gotten to the point that right-wing groups are pushing US legislation to outlaw it as “terrorism” [1], with the president of the right-wing American Heritage think tank pointing to examples such as “college students who took over campus quads” [2] and town halls with congress members getting shut down [3] - they say new laws are needed because public anger over “political purposes” is disturbing “civil tranquility” in the U.S. so those who are aiming to impact “civil tranquility” need to be “held accountable” under the law.
The legislation being pushed starts with:
> Section 1. Short Title. This Act shall be known and may be cited as the “Combatting Civil Terrorism Act.”
> Section 2. Purpose. To promote civil tranquility by holding those who seek to disrupt it for political purposes accountable.
And aside from that, social media (Reddit) seems awash in “public anger” etc.
[1] https://sapirjournal.org/fixing-america/2026/islam-in-americ...
[2] https://manhattan.institute/article/model-legislation-an-act...
[3] https://www.city-journal.org/article/adam-smith-israel-civil...
I expect that most voters welcome efforts that aim to stop other groups using violence and intimidation to suppress their speech or legal religious activities.
The only problem with these proposed laws is the same problem with all existing laws - selective enforcement.
There is a lot of public anger tho. Facebook and Twitter are working overtime to stroke it. Heritage loves it as long as it ends up being in MAGA colors.
I agree with you, that public anger still exists[1], but I will not consider that link I mentioned in my response as an example of the suppression of public anger.
A new law that specifically calls out people as criminals when they attempt to limit a person from entering their place of worship, etc is not a law suppressing the expression of public anger.
======================
[1] One can even make the argument that Trump got elected on the back of public anger. The values he espoused are unpalatable, but there is little question that the whole suppression of opinions in the last decade have had much to do with his popularity.
I can see a scenario where companies like openAI and Anthropic do go belly up but the technology and IP and physical assets remain, get balkanized or snapped up by various other parties. Let's say for instance that they do finish the physical construction of the "Stargate" datacenters in Texas, and they get filled with the equivalent of a whole shitton of B300 RAM/GPU systems and are up and operational. Those don't disappear.
In some kind of catastrophic failure scenario it could end up as a debtor in possession arrangement, or chapter 7 sale to new set of people who want to make use of it. Not unlike what happened to a number of much smaller scale datacenters that were built with 1996-2000 dotcom 1.0 boom money that changed ownership around the 2001-2002 time frame.
Unfortunately, that's not the case. Between the big 5 (Microsoft, Meta, Amazon etc.) they're spending more than $600 Billion in 2026! They don't have that much cash lying around so they're selling bonds!
That's debt!
Not only that, they're increasing the bond sales in Europe! I assume that means they're tapped-out in the US!
All this off-the-books stuff, despite being legal but shady, is still debt! Debt has to be paid by someone.
To sum it up: the AI buildout is a highly leveraged, debt-fueled expansion, not an organic, cash-funded software cycle... this will not end well!
Much of the dotcom era telecom expansion depended on debt and projected future demand, but the resulting networks were so overbuilt that only about 2% of North American long-distance capacity was being used. Falling prices left numerous operators unable to service their debts. Today’s AI expansion is led mainly by already-profitable companies with large existing revenues and cash flows. Microsoft alone generated $136B in operating cash flow during its 2025 financial year while spending $65B on capex. It held $95B in cash, compared with total debt of about $43B. That means a disappointing return on AI investment would reduce profits, cash reserves, and shareholder value, rather than making these companies insolvent.
It's the pure AI companies like OpenAI that will hit the wall.
However, the tech companies are increasing their debt enormously. That's the issue.
There isn't a winner and loser.
If you loan me 10 pounds and I simply keep it or spend it but refuse to pay you back - aren't I the winner and you the loser?
Claiming that the market is a positive sum system, because I use that 10 pounds to invent something that changes world productivity is missing the fact that there is a constant stream of huge energy inputs from the sun - that's ultimately what allows the local entropy to decrease - not that you lent me 10 pounds.
And that's complex - even in your simple example you can argue that people have likely improved their capability to dig and fill holes - and the latter skill is particularly valuable in the UK right now.
ie Entropy includes information component :-)
It's not fighting against tyrants, it's a declaration of an intention to be the next tyrants.
Really changed so little in my life. I just use chatgpt instead of Google search now
It's being marketed as some ground breaking revolutionary change when all it is is a better widget
Remove it and the worst that will happen is that I'll go back to writing my emails and my code by hand - something I've done for the last 20 years anyway
In fact, bank failure and then having the government only guarantee ~$50k of funds per person is a good way to hand wealth to the people and take it from corporations and the super wealthy.
Hence the power that be try to prevent it best they can. In some cases, unsuccessfully (Iceland) and in others, a bit more successful (the US). But seemingly, they're only able to delay the inevitable, not completely prevent it.
What you suggest will hit normal persons.
What we really need is people going to jail that let stuff like that happen. Let them bear actual responsibility for their 1,000k+ salary.
The global economy isn’t pivoting off Icelandic banks the way it is off US ones.
I was going to write a shitty reply to this, but the more research I did, the more it seems actually this went pretty well for them. UK and NL governments that lost deposits in the Icelandic banks mostly got their cash back (eventually), there was recession and unemployment but not that much worse than other countries, and the country is in good standing again with the markets.
I would note that this is much much easier to do if your investors are foreign, rather than domestic pension funds, so it doesn’t bring down the government, though.
If a bank like JP Morgan were to fail, the event would be without exaggeration cataclysmic to everyone, even small local banks and credit unions. Even if ultimately the clients of JPM could be made whole, the weeks of uncertainty and frozen funds would single handedly obliterate the financial systems across the globe. It's the age old adage: "if you get margin called and get wiped out, it doesn't matter that you'd have recovered just 2 weeks later". The world as a whole is deeply leveraged. It's that leverage that affords us the ability to supercharge all the growth, from AI to drugs research, insurance, EV... everything. But that leverage comes at a cost, which is that systemically important institutions failing can have disastrous cascading de-leveraging effects.
I'm sure we'd more or less all survive and the world would recover, but it would likely be a 2008 GFC style scenario most likely.
Private credit is not supposed to be systemically important and it's not supposed to need bailing out. Maybe we'll find out how true that is in practice.
We can check if governments went back to their own tricks or not after 2008. What were the interest rates after 2008? Quick check ... yep, they went back to the old tricks.
When "shit hits the fan". Either governments really cut spending, leave pensioners without income and see their new loan interest spike to 10%+ (if they're lucky) ...
OR they do a massive cash injection saving the banks, and increase spending.
Which would you chose? It doesn't even matter. If you were the government and chose the first option, you would quickly find suddenly everyone across the political spectrum uniting to depose you. A decent chunk of people would literally not have any other choice but to do that.
(of course, as per usual governments aren't behaving "correctly" according to economic theory. If you take the old "government as spender of last resort" governments should have radically cut spending in the last 10 years, because there was no need for extra spending. If you take the new "government invests pensions", then of course government spending needs to be investment. In other words, governments only allowed to spend like a normal investor, expecting and demanding a return. Ie. no social spending increases. In practice governments increased spending at any cost, and so we're in trouble again)
Oh and should I mention that for any other entity investing pension money in loans to yourself is a very unique thing in law. It is THE ONLY financial crime where corporate structures do not protect management. You're the worldwide president of ExxonMobil? Doesn't matter. You do with employee pension money what government does with it? (ie. loan it to yourself, in this case ExxonMobil). You go to jail. Directly and long-term. In America. In Spain. In Japan. In fucking Vatican city.
Two concerns:
1. In a duopoly it won't doom either one party, not if the bailout-seekers involved are able to bribe them both equally.
2. A politician in not-so-bad party could let the other party doom itself in a long-term diffuse sense... Or they could accept some generous short-term very concentrated-on-me donations. The interest of the group is not always the interest of the members.
The US desperately needs voting-reform, to reduce the spoiler-effect and winner-take-all mechanics.
I honestly can't tell if you're talking about the Right pandering (in words only) to the voters or the Left scoring own-goals in the last election?
One could only hope. In any case, in the short term, with AI having such a poor public-image story, if the Dems want to get back into power again they'd do everything they can to assure their voters that a bailout is not on the cards.
This is the first single-issue voter concern that has really popular support, and they'd be dumb for chasing identity politics again.
But the latest election made me doubt that is actually their goal.
It's one of the many reasons that partisanship is foolish - it essentially leaves governments unaccountable. Another is the reason I put "they" in quotes. There will be lots of partisan finger pointing when the bubble pops, but it was both the DNC and GOP that bailed out the banks, and it'll be both of them bailing out big money again.
That gambling actually led to people getting those homes. The people losing them was the correction.
> the electoral effect was basically zero.
because...
> it was both the DNC and GOP that bailed out the banks
Look at who got elected, all those scandals, consistent lies, empty talk and fraud yet support is still strong.
That's how revolution starts.
And there are many revolutions that we (in USA) claim are good! Our own, from Britain, the French one. Of course others are bad (depending on where your vantage)
It's just that, sometimes the murder is (retrospectively) justified. Like in a war, they're killing us so we kill them.
But class war isn't fought with uniformed combatants.
And, in USA at least, many of our politicians are bought quite blatantly. If they aren't serving the citizens perhaps they have thrown out civilization and the social contracts first?
It's all quite complicated.
In general though, murder bad and government should serve the poorest citizens forst, corporations last.
I think it's a real pity that of the "four boxes of liberty", Musk has interfered with the first three: soap, Twitter; ballot, the lawsuits about million dollar rewards for voting; jury, selecting his jurisdiction, dismantling enforcement institutions via DOGE.
On the plus side, it's only "interfered with" rather than "completely eliminated".
Still, glad I have an ocean between me and the USA.
And the importance of the justice system working properly is that when it fails constantly and deliberately, many will come to the conclusion that the only option left is murder.
I suspect the real profit is made by also being in a position to adjust laws and rules to make sure your investment survives, as well as turning low rated bonds effectively into government bonds. All those things are at the expense of others in the bigger picture.
Aren't the models from Anthropic and OpenAI simply the distilled work of everyone else who ever put their work online, or in books?
Why is their distillation okay, but other distillations not ok?
Nope, seems I'm past the edit window. Oh well.
What is the economic difference? LLMs have been trained on the results of billions of dollars worth of time, research, investment and expenditure. When you ask an LLM a question, they are giving you the results of those billions, or hundreds of billions, effort.
Those things weren't free; they cost money to produce! If anything, the OpenAI and Anthropics of the world got more economic value for free than the people distilling them did.
World's smallest violin.
Also, how a model identifies itself isn't very telling, many models when asked in Chinese will identify as DeepSeek
By your own logic Anthropic must have distilled from Chinese models rather than produce their own Chinese training data.
Here is sonnet acting like deepseek: https://x.com/stevibe/status/2026227392076018101
And if you ask Opus 4.8 in the API: '你是什么模型' (what model are you?) It responds ~9/10 times with:
我是通义千问（Qwen），是阿里巴巴集团旗下的通义实验室自主研发的大语言模型。我可以帮助你回答问题、创作文字（比如写故事、写公文、写邮件、写剧本等）、进行逻辑推理、编程、翻译等等。
有什么我可以帮你的吗？
(I am Tongyi Qianwen (Qwen), a large language model independently developed by Tongyi Lab of Alibaba Group. I can help you answer questions, create text (such as writing stories, official documents, emails, scripts, etc.), perform logical reasoning, program, translate, and more.Is there anything I can help you with? )
I can't really judge how I might personally benefit from the current direction because I am uncertain what that direction is, and indeed if it will be sustained.
If you are trying to allege that I have some personal affiliation to an organisation where I might benefit more than any average member of the public, then I must disappoint you. I work on my own projects, I sometimes make money from games that I make.
I do not wish for a hypercapitalist dystopia where a company could train a model on work for hire content and then forever churn out content that they assert ownership of. I don't want the suppression of ideas caused by others asserting that because they had an idea, they own it and can dictate who gets to use it.
I also do not wish for another reign of terror.
I don't think violence will make the world better, and I don't think disaffected chique will either. I advocate identifying problems, suggesting solutions to the problems and working with people to reach consensus to apply those solutions. I assume that humans are fundamentally good and in the absence of religion can generally be reasoned with. Even those who do things you disagree with probably are doing what they think is best.
I may be wrong about the goodness of humanity, but it does not matter, If it exists it's worth working with. If it does not exist, it is not worth saving. I choose to assume that it exists on the chance that it might.
My use of it took the form of
A statement on how a guillotine kills people when they are powerless.
The negation then, in the first part compares that to the connotation of the previous post of an underdog battling those in power, and rejects it as inconsistent and the negation provides an alternative framing that is consistent with the action of killing a person who can no longer harm you.
Both sides of the negation make a point.
This is obviously much more common in a dialogue than a monologue, because the first part is used to reference words of the other party. In a monologue, that other party is the past self of the speaker, it suggests a change in position. I think models end up using as a short hand for realisation.
Are there any good collections of negation usage in AI with the surrounding paragraphs? I'd be interested to see what percentage of them are an expression of self realisation.
Also, there is the issue of human language taking some 'tells' because they are exposed to LLM output a lot.
I'm still grumpy that LLMs made it nearly impossible to use the words 'delve' or em-dashes without getting called out as an LLM.
Fortunately, you keyboard warriors lack the skill, connections, and charisma to actually enact any of your bloodthirsty fantasies.