This reminds me of around 2002 when I wrote an article looking at how all the web behemoths at the time were claiming profitability through ad sales, but actually the vast majority of ads were from one web behemoth advertising on an other's site and vice versa.
This is How Larry Elisson beat Musk for a short while although Oracle was struggling.' The deals among Oracle, Nvidia, and OpenAI have raised concerns about their circular and potentially risky nature. Oracle is reportedly spending tens of billions on Nvidia's advanced chips, Nvidia plans to invest up to $100 billion in OpenAI, and OpenAI uses Oracle's cloud infrastructure through Microsoft's Azure, creating a closed loop of financing and business. Some analysts warn this creates a reflexive loop or "dangerous bubble" where valuations may be inflated artificially by the circular flow of capital and resources, reminiscent of past booms that ended in sharp market corrections.
You can't discount the impact this will have on global stock markets and what that may do to both individuals and, more importantly, pension funds, as a large swath of people are retiring
Damage is unavoidable, we can only hope that it happens to someone deserving. Innocent individuals and pension funds still have time to retreat; if they don't it will be their fault.
We are moving towards gerontocracy - if pension funds will have large losses it’s very likely that young, working age people will be taxed extra heavily to keep the QOL of pensioners.
That would likely lead to a revolution: Millennials are 30-45 and they’re not in a good place; neither is Gen Z.
We’re already seeing revolutions elsewhere — and it’s likely that trying to loot them further by generations who sold out the nation will simply lead to social collapse.
Have you ever battled an 87-year-old wearing mechanized battle armour? They're crazed, hopped on speed, eyes goggling in their sockets
A pack of 3 oldies burst through our perimeter one winter night... the screaming woke me up. Outside my tent the forest was lit up red by our laser blasts, trying desparately to take them out.
We thought that a revolution would be a good idea, but an upside-down population pyramid is a hell of a thing when you're on the bottom.
Yes but this is by design. Wealthy people want bubbles, when they pop, you can gobble up all the value at all time lows. Then you hold until you don't feel like it anymore as the market goes back to growing. You see this as how private equity has bought up real estate across the USA to turn into rentals after 2008, for example.
Aren’t extremely wealthy people that wealthy due to the valuation of their stock? IIRC generally the higher the networth, the higher share is kept in stocks
And they don't spend money, they take debt against their existing assets to fund projects and investments. So long as they can service the loans across economic downturns, they don't particularly have to feel the effects of a recession, outside of the mentioned opportunities to buy the market at a discount.
I suspect $$ is just a number for them. Being able to control more resources is the ultimate game. You gotta have zillions of $$ to join the tournament, though.
Person A has a net worth of 2B
Person A has a loan at 500M backed by their holdings
Stocks drop 50%, Net worth is now 1B
Person A buys $500m of stocks
Market Recovers 100%
Person A now has 2B original holdings and 1B gains, $500m owned = 2.5B
Very simple example, and not the only way to do it - but people need to remember net worth being 500B is not 500B in the bank, and at some point the number doesnt matter
More importantly you keep the portfolio semi-balanced.
Just using Google / Gold as a comparison [1].
Assume you have 100 units of each.
In late 2021, Googs gone up ~100% so you have to rebalance because you have $200 in Goog and $92 in Gold. So lets say you rebalance to 80 Goog (160$) and 144 Gold ($130).
In late 2022, Googs gone down ~40% so you have to rebalance because you have $96 in Goog and $141 in Gold. So lets say you rebalance to 100 Goog ($120) and 118 Gold ($112).
So over the course of 2 years Goog has gone up 20% and Golds gown down 5% but your investments are overall up 16%. Obviously a 100% Goog investment is higher but with more risk.
If you didn't do any rebalancing then you have a gain of 7.5% (100*1.2 + 100*0.95 = 215)
There was an article here recently about alpha school that mentioned this is how of of its backers made a lot in the aftermath of the dot com bubble - picking up companies with products that happened to implode in the bust. He's not the only one and that's not the only bubble that had that happen
> If this means that the fall of OpenAI will cause Oracle and Nvidia to crash and downsize
Only if those are the only actual real customers. It's not. The pie is a lot bigger. And with the current hype some other AI company will just take over and the bubble will continue.
A shareholder going bust is a no-op. However, the loss of large customer, is concerning, but less so, relative to the market souring on AI investments as a whole, which would be not great for AMD, but disastrous for Nvidia.
I doubt, or at least hope, that no sort of managed fund for retirement (read: low risk) is going 15% on AI stuff. Regardless of what one thinks might happen, it has to be accepted at this is an obviously very high risk bet, because this all is screaming bubble even more than the era of 'zomg i flipped this house i financed on an insta-approved loan, installed granite counter tops, and walked away with $20k profit in 4 days.'
Failing to diversify is the fund managers' problem. Pension funds shouldn't have all their investment in a single countries' stock market (doesn't matter that most, if not all, of those companies operate worldwide, they are still from the US), not even in equity markets. A wiser investment manager (or individual investor) would put a chunk in different kind of equities, and diversify into fixed rate, precious metals, perhaps real estate... Not bet it all to the volatile S&P 500. "Past performance does not guarantee future results.".
If it’s a bubble, then the long term value people are now counting on to retire with in the future is not actually real though, right? Like a bubble popping doesn’t destroy real value that used to exist, it exposes fictional value that never really existed.
What’s the mechanism you have in mind for that happening via bubble? They sell early before the crash? I’m curious if that actually happens, since the mass psychological nature of market crashes seems to make them difficult to predict reliably, even for people with a lot of resources.
Inflation on the other hand definitely does reallocate real value to whoever already owns a lot of assets, and I believe inflation often goes along with bubbles, since credit expansion is what usually kicks off the bubble in the first place.
Is it fair to say OpenAI is in a sense “washing” the money passing between Nvidia and Oracle? And instead of taking a cut in the traditional money laundering they are enjoying massive valuation gains?
When the fed investigates this does it matter if one of the 3 companies is not a publicly traded company?
I guess you could see it like that but I don’t think it’s “wash trading” in the sense that they’re coordinating it. I don’t really know what the fed would investigate here.
What you're missing is that real value can be created in these exchanges - Nvidia really makes chips for example.
The issue is that it's an industry of investment which exists solely to power more investment in AI - the entire chain is still assuming that someone will eventually pay for this.
At the end of the day all that money leaks out to employees and suppliers...but no one those people transact with may have any interest in buying what was produced.
That's only really value if the chips are useful and if there are people buying the chips for something they want to do with them.
It's entirely based on the perception that LLM training & inference is here to stay at ever growing scales when the shortcomings of Artificial Dreaming are increasingly scrutinized. Not all businesses want to end up paying refunds to their clients like Deloitte [1] because the LLM hallucinated crap into their reports (and they failed to correct it).
> It's entirely based on the perception that LLM training & inference is here to stay at ever growing scales when the shortcomings of Artificial Dreaming are increasingly scrutinized. Not all businesses want to end up paying refunds to their clients like Deloitte [1] because the LLM hallucinated crap into their reports (and they failed to correct it).
This assumes Deloitte didn't make more $$$ from the deal by "outsourcing" it to AI than not. It was a partial refund.
> You haven’t brought in the side costs into your analysis
And you haven't considered that Deloitte might get this money some other way anyway. It's been budgeted and needs to be spent by the government department for this financial year. They'll find another project to hand out.
That's very cool. I was going to get a hetzner for some hobby projects.
Can you get that without adding a payment mechanism? For some reason, I'd be more cautious about putting my card into Oracle's offering than anyone else.
> Can you get that without adding a payment mechanism?
No!
There are actually two free tiers.
The limited free tier you get without adding a CC and then this advanced "always free" tier I am using, where you have to use and verify a CC and have to pay big bucks if you mess up.
So it's free... Until is not. My take on this would be that if Oracle makes any change to that "free" tier and you do not take proper action you might be automatically charged hefty amounts from your credit card. No, thanks. It's way way safer to spend few bucks a month in other VPS options than get a big surprise on your credit card.
I know Oracle are notorious for scummy license enforcement, but in practice their Always Free tier is superior to Amazon's precisely because they will cut off your service rather than bill you arbitrary (huge) amounts.
The big limitation is the 10mbit egress bandwidth, but you can work around that with (free) Cloudflare.
I think the point here is that free beats easy peasy, especially when learning a new skill, where your easy peasy isn't their easy peasy. And getting something up and running on Oracle Cloud with Terraform is nowhere near easy peasy for someone who never did anything with Terraform ;)
Even smart people belive a lot of untrue things, and his domain of knowledge isn't likely to include much more biology than my GCSE grade C and a handfull of brilliant.org micro-lessons; it's quite possible, though I would not put specific odds on it, that he's seen e.g. the sudden change in the rate of progress in protein folding, where before it was ~ one PhD thesis per protein and then suddenly AlphaFold did all of them, and extrapolated from that to everything else like an over-enthusiastic PopSci reporter.
(Interestingly, some of the world's dictators do seem to have an interest in the current state of the art in prolonging life. For example Xi and Putin chatted about organ replacement https://www.bbc.com/news/articles/cr70rvrd41ko)
Ah yes, my favourite philosophical dilemma, The Organs of Theseus, and whether or not a sufficiently organ-replaced person is absolved of their previous actions by virtue of not necessarily being the same person.
As a more serious point related to this though, I was under the impression that organ replacement didn't address issues with telomere length?
> What's he going to do with all that money, and what does he care for the risk it's bad or shady?
The thing that unites Ellison, Trump, Musk, Thiel and a fair few of the other politically active billionaires is the obsession with "legacy" - they want to leave their mark in the history books, figures which will likely be remembered and taught in schools in thousands of years similar to Roman emperors.
Musk is the most obvious with his obsession of settling (and eventually dying) on Mars, Trump is dreaming of getting a Nobel Peace Prize (if only to not let Obama be the only US President who got one), and the rest is hunting for the "invented AGI" crown.
Obama was the fourth U.S. president to be awarded the Nobel Peace Prize, after Theodore Roosevelt (1906) and Woodrow Wilson (1919)—both of whom received the award during their terms—and Jimmy Carter (2002), who received the award 21 years after leaving office. - wikipedia
What does Ellisons personal wealth have to do with this? The concern is that the circular pattern of shifting money between these companies is artificially inflating the stock market to heights that will crash very very badly when this bubble finally pops.
>By 1998, Yahoo was the beneficiary of a de facto Ponzi scheme. Investors were excited about the Internet. One reason they were excited was Yahoo's revenue growth. So they invested in new Internet startups. The startups then used the money to buy ads on Yahoo to get traffic. Which caused yet more revenue growth for Yahoo, and further convinced investors the Internet was worth investing in. When I realized this one day, sitting in my cubicle, I jumped up like Archimedes in his bathtub, except instead of "Eureka!" I was shouting "Sell!"
That's not how Ponzi schemes work. Yahoo had a defacto _monopoly_ and the market had bad discovery leading to bad price information. There was no point at which the internet was /not/ worth investing in and everyone who had experience with it knew that.
The real problem seemed to be that you can only put so much money into pets.com before it becomes stupid. You had more short term investment capital than could be _effectively_ spent at the time. The long term players, as usual, avoided the Archimedian idealism, and were heavily rewarded anyways.
Same issue with LLMs - there probably is real value there, but investors have rammed 10 years' worth of investment into the market in 10 months.
That means:
1) Late investors end up taking a bath because it will actually take 10x longer to get a return than they thought.
2) Investment becomes inefficient - there are lot of GPUs being bought in 2025 that should have been bought in 2030. By the time 2030 comes along, those 2025 GPUs will be outdated, so the value they will provide will be less than if they had been purchased at the correct time
>There was no point at which the internet was /not/ worth investing in and everyone who had experience with it knew that.
I don't get this. Of course there was. Specifically, March 10, 2000.
To declare that something is not worth investing in doesn't imply that it's useless. (and actually, swing investors will argue something can be worth investing in even though it is in fact useless, e.g. web3 and most crypto).
Why is it so popular to "akshually, Ponzi schemes are different!" these days?
Words have meaning in context, and calling something a "Ponzi scheme" these days means any pyramid-like system that only works as long as new investors come to the table, because that new investor money is the sole source of gains for earlier investors. But once you run out of new investors, which is inevitable, the whole thing collapses. What pg described was exactly a Ponzi scheme in that sense, even if it wasn't a deliberate scam. And that's very different from other types of business ventures that eventually fail because of the more "normal" reason that they just don't gain traction.
And you do recognize the context of those words was 15 years ago at this point?
> was exactly a Ponzi scheme in that sense, even if it wasn't a deliberate scam
There are better terms of art to describe the scenario. If we lacked for those words I might grant you this point, but since we don't, I find the description lacking in precision if not completely faulty.
> the more "normal" reason that they just don't gain traction.
Is that actually the "normal reason" most businesses fail? I'm not sure it is.
Agree, people became extremely pedantic with words and it's incorrect to assume they only mean what is described in a dictionary, as if we are in a court.
Language is fluid, and many of words got their very known & popular meanings later, when people used for things that it didn't exactly mean what it was initially intended to, and it got popular.
People are ever less empathetic and sociable, this being the reason for such comments.
People become pedantic when it's the only way to defend themselves and their friends or to argue against someone.
In this case dotcom investing was so stupid that it should have been recognized as stupid at the time, without hindsight.
For someone who lost money, gave bad advice etc. insisting that they didn't fall for a Ponzi scheme means representing themselves as less terrible investors, regardless of facts.
But you can't fake profits that way (without cooking the books); you can only fake revenue. That's why profits matter.
Though I guess with high velocity of money circulating in tight loops, everyone involved can feel rich. 1 million dollars changing hands once per day can potentially mint 365 'millionaires' during the course of a year. If they just pass it back and forth among themselves to buy each other's stocks and products and don't let that money escape their network to actually pay for something useful, they can all be millionaires... On paper.
Not to mention how relatively small amounts of money, moving at high velocity, can inflate asset prices... On Alice's birthday, Bob can buy 100 shares of Alice's shell company at a price of $10 per share. If Alice owns 1 million shares of the company (which she founded), it means that Alice's net worth is at least $10 million... It cost Bob only $1000 to give Alice a net worth of $10 million. Now imagine that same $1000 moving in and out of that company's stock, traded 1000 times per day between various people. With just $1000 circulating back-and-forth at high frequency (A.K.A. high-frequency trading), you can generate a trade volume of $1 million per day... So it all seems reasonable; a company with a $10 million valuation with $1 million daily trade volume... Nothing suspicious. So you can basically start a shell company and fake everything; from the market cap, to the trade volume; using only a relatively small amount of money. This is what they were doing in the early days of crypto.
The same money can hop around in circles between an insane number of people when it's just 'revenue' because revenue is not taxed (after expenses). Only profits are taxed... But the same money, as profit, can barely hop between 6 people before it's taxed down to just 10% of its original number. High-velocity revenue can severely distort people's perceptions of the market, especially in the era of media filter bubbles.
Damn. Didn't think of that. Seems I have a very naive definition of 'profit'.
It's crazy how all these small subtleties in definitions of various terms can create something so abominable in the aggregate. Makes one want to reach for a tinfoil hat. Not sure I can even trust the company which makes the tinfoil at this point.
In the very short term yeah, but in the long term GAAP still sniffs this out. Eventually Company A will need to use some form of mark to market valuation to assign a value to that $1M they invested. If Company B is a dumpster fire then eventually the $1M investment will be worth much less (or zero) and balance is restored to the world.
You see companies hand waving this sort of thing away on earnings calls when they have a really bad quarter but the CEO is blowing smoke at shareholders saying “On non-GAAP this was an amazing corner.” Which is code speak for “that terrible decision I made a few quarters ago is coming home to roost and our lawyers have said I need to reflect that on our books. If you ignore where we screwed up badly, our numbers look good!”
> it means that Alice's net worth is at least $10 million...
Some people might interpret or represent the aforementioned transaction as justification for thinking Alice's net worth is at least $10M, but it is not an objective measure of Alice's purchasing power (as it would be in the case that Alice had $10M cash).
>The same money can hop around in circles between an insane number of people when it's just 'revenue' because revenue is not taxed (after expenses). Only profits are taxed
On the US federal level. Not sure about other countries, but quite a few other governments in the US do tax revenue.
>Nevada, Ohio, Texas, and Washington impose gross receipts taxes instead of corporate income taxes. Delaware, Oregon, and Tennessee impose gross receipts taxes in addition to their corporate income taxes. Some localities in Pennsylvania, Virginia, and West Virginia likewise impose gross receipts taxes, which are generally understood to be more economically harmful than corporate income taxes.
All clearly designed to cash in on the speculative value being given to AI companies.
Strategically I thought OpenAI's deal of getting 10% of AMD for driving their stock price to $600 was a pretty clever way of creating $97 Billion from nothing - effectively paying for the GPUs they'd purchase.
At the same time I would've thought this insider pump and cash-in strategy would be somehow illegal, but I guess anything goes with this administration.
These sorts of shenanigans happen under whatever administration happens to be in office at the end of a long business cycle and near the frothy peak of a hype cycle. And that frothy peak can be quite tall and last for years.
Just getting options and trying to get the company to do well isn't illegal or discouraged. Price manipulation can be illegal but I don't see any evidence of that so far.
This is going to end badly — and soon. It’s ironic that catastrophic forgetting (i.e., the inability to perform continuous learning) and hallucinations (i.e., the failure to recognize when a prediction is unfounded) won’t be the causes of the crash, but rather greed and stupidity.
I finally got around to watching that a year or two ago. That scene is among several that are just remarkable bits of work, but somehow the whole film itself felt like a slog.
I feel like that's the problem generally with late career Scorsese. The last of his films that felt consistently interesting and free of flab was, to me, Casino, though he got close for me with The Aviator and The Departed.
Agreed on TWoWS. Kind of loved every scene in the movie, but actually strongly dislike the movie in full. I also think that despite what seems like an attempt to critique the destructive lifestyle, it does the opposite and glorifies it. Mostly not a fan of his later career work, but I do really like Killers of the Flower Moon and Shutter Island.
I'll be honest and say I never saw Shutter Island in part because I read the book (at one point I was a Lehane fan) and saw the twist coming miles and miles away.
I haven't seen KOTFM b/c it's past my Scorsese Endpoint. ;)
Anecdotally, many people at OpenAI, Nvidia, Oracle, etc., sincerely believe their company's own hype.
They remind me of the story about the oil prospector in Warren Buffett's year-end 1985 letter to Berkshire Hathaway shareholders:
> An oil prospector, moving to his heavenly reward, was met by St. Peter with bad news. “You’re qualified for residence”, said St. Peter, “but, as you can see, the compound reserved for oil men is packed. There’s no way to squeeze you in.” After thinking a moment, the prospector asked if he might say just four words to the present occupants. That seemed harmless to St. Peter, so the prospector cupped his hands and yelled, “Oil discovered in hell.” Immediately the gate to the compound opened and all of the oil men marched out to head for the nether regions. Impressed, St. Peter invited the prospector to move in and make himself comfortable. The prospector paused. “No,” he said, “I think I’ll go along with the rest of the boys. There might be some truth to that rumor after all.”
>And so, you really
want every person to be able to have
their own dedicated GPU, right? So,
you're talking order of 10 billion GPUs
we're going to need. This deal we're
talking about, it's for millions of GPUs. Like, we're still three orders of
magnitude off of where we need to be. So
we're doing our best to provide compute
availability, but we're heading to this
world where the whole economy is powered
by compute... (15 min in)
Rewatching the video it strikes me that there is no talk of how the compute will be paid for. Altman is saying there is so much demand for compute but a lot of that is because he's setting the price for users to zero in spite of the service costing a lot to provide.
I guess they figure once they have the users they'll monetize somehow but that bit's kind of iffy.
I agree, though it's not out of the question that there could one day be a single country whose economy relies entirely on data centers, similarly to petrostates. Maybe we'll call them datastates, or compustates.
If their compensation includes stock options, they have no choice but to believe. At Enron many employees went as far as to put 100% of their savings into buying the stock.
This incest investing is pretty telling. The whole economy is about to get burned- doubly so with DJTs tariffs and bonds not looking like the safe haven they usually are during turbulent times.
“We’re now locked into a particular version of the market and the future where all roads lead to big tech,” says Amba Kak, co-executive director of the AI Now Institute, which studies AI development and policy. Indeed, the success of major stock indexes—and perhaps your 401(k)—is resting on the continued growth of AI: Meta, Amazon, and the chipmakers Nvidia and Broadcom have accounted for 60 percent of the S&P 500’s returns this year."
I notice that when I ask ChatGPT a question the answers I get back seem more verbose than they were a year ago. Where a 1-2 line response would be sufficient GPT delivers a sprawling essay. Overall the value of the answer has probably gone down, but tokens are up.
You can quite easily ask it to summarize the result in a sentence or paragraph. LLMs have no other way to compute than write text and the more text they write the more compute they do. You only care about the final output.
SBF lacked funds to ride it out. If they survived for a bit longer, Bitcoin would have surged and they'd keep going. AMD, Nvidia etc all have income and funds to survive.
This is just about finances unless you're implying there's some crime here too.
For an interesting interpretation of the recent AMD-OpenAI deal, see Matt Levine's column from a few days ago:
> OpenAI: We would like six gigawatts worth of your chips to do inference.
> AMD: Terrific. That will be $78 billion. How would you like to pay?
> OpenAI: Well, we were thinking that we would announce the deal, and that would add $78 billion to the value of your company, which should cover it.
> AMD: …
> OpenAI: …
> AMD: No I’m pretty sure you have to pay for the chips.
> OpenAI: Why?
> AMD: I dunno, just seems wrong not to.
> OpenAI: Okay. Why don’t we pay you cash for the value of the chips, and you give us back stock, and when we announce the deal the stock will go up and we’ll get our $78 billion back.
> AMD: Yeah I guess that works though I feel like we should get some of the value?
> OpenAI: Okay you can have half. You give us stock worth like $35 billion and you keep the rest.
I would be convinced that he's some sort of super well calibrated financial article AI engine (due to consistency of output and breadth of coverage), except that I've been reading his column long before LLMs were on the scene.
There's no way someone would be stupid enough to part with $x worth of product and $ 0.5 x worth of stock in exchange for $x, is there? Honestly, it's stupid to even make such an offer. What prevents AMD from turning it around on them?
> AMD: Oh, since you put it like that, why don't you pay us $x for the chips and give us $ 0.5 x worth of stock? When your market cap rises by $ 0.5 x you'll get the money for those stocks. And hell, it might even rise further, giving you back some of the money for those chips. You're really robbing me blind with this deal, you'd be a fool not to take it!
>What prevents AMD from turning it around on them?
Matt's telling is a little facetious, so the exact negotiation probably didn't go like that. More importantly, OpenAI has better options. If they want to pay AMD cash for chips, they could do that without also giving up their equity. The same can't be said for AMD. They're the second choice when it comes to GPUs, and they're desperate for market share, so they're willing to cut deals like this. Without this deal OpenAI probably would have bought Nvidia chips with cash.
AMD might do it just to try and be (stay) relevant in the market. Prior to this, discussions about AMD were slim to none, "just another inference card manufacturer". Now they're shining in the spotlight. Mission accomplished.
If you really think about it though, inference is the end goal for most AI models. So I don't see "inference only" being necessarily bad, especially if you get more bang for your buck.
Because OpenAI has the hype and delivers the pop, not AMD. An AMD investment in OpenAI would probably have had a similar effect on AMD, not an equivalent one on OpenAI. (Either way it's Ooh AMD is getting seriously into AI, awesome, suddenly it seems worth 100x whatever it was before!)
Replace AMD & OpenAI with some crypto company and Trump, hypothetically the latter doesn't really obviously get any significant benefit from the former's endorsement.
Oh this just seems like the typical financial dealings that City of London invented to obfuscating ownership of holdings half millennium ago. Some would event argue this obfuscation what transferred power from the military to bankers. And it is how we end up in this transnational corporation where the public really have no idea who is owning what and who is actually running the show here.
Now we are just transferring the power from bankers -> tech/AI.
This reminds me of the AOL Time Warner merger. "Back in the day", America Online (AOL) had a valuation large enough to pull it off:
It was the largest merger in history when completed with
the combined value of the companies at $360 billion.[0]
After less than a decade, once ISP's became a thing, the AOL part of the new company shriveled into nothing and was spun off. It still exists, in name if not anything recognizable to what it once was.
Does he explain why the valuation of the company would go up by $78 billion on the public news that AMD agreed to give $78 billion of goods to OpenAI for free?
It validates that MI450 works. Jensen was saying other GPUs are so bad that companies wouldn't even take them for free and then OpenAI said no, we actually will take AMD GPUs for free. They're not that bad.
It’s just AMD levering up on its on stock. Similar to MSTR. AMD is taking a loan against its own shares to build gpu’s hoping to rent them out via OpenAI in the future. No point in betting against it, because it can’t get margin called until OpenAI goes bankrupt. Which isn’t happening until OpenAI is worth $10 trillion first.
From the OpenAI side, it’s an amazing deal. $35B profit in one day… at a pe of 30, it’s already worth a trillion if it can repeat similar deals every year.
People are bringing up the AMD deal but isn't that giving it too much credence? The deal hasn't had any material consequences yet apart from stock market fluctuations. The bigger problem for me is that AMD doesn't seem like is going to be a player of note in the AI sphere. So this deal like many other big money AI deals look like optics to me.
> AMD doesn't seem like is going to be a player of note in the AI sphere.
I thought AMD is well positioned in the inference space? They have high VRAM, somewhat high connectivity, already shipping pods that are pretty ok for inference? Training is still dominated by nvda and their CUDA moat, but there's an increased need for scalable inference, and that should fit AMD well. Am I wrong in thinking that?
The MI450 is the pets.com from the 90s of AI hardware. It's bad.
The bubble now isn't even about LLMs, but has affected the entire stack. Even hardware companies are offering rubbish for the sake of prop'ing up their own valuation.
CUDA had decades of improvements, we may have AGI (tm) earlier than what it will take AMD to fix their broken software.
MI450 doesn't work well at all, OpenAI would rather take the deal and overspend in a project to try to use it for their software stack, which is incompatible with it.
Just for the sake of proping up valuations, with no "market competitiveness" guidance. Just for moving money.
Funnily enough, Google is ahead of all of this, as they can make their on TPUs and have their integrated stack. They don't need to generate fake economic activityu.
On the surface, the circular financing seems worrying but once you dig into it, it seems quite benign, tbh. NVIDIA is handing out GPUs into OpenAI, in return for stock/future revenue, whichever way you want to slice it. OpenAI has wonky unit economics, yes, but they're growing at ~100% YoY, so it all checks out, if you ask me.
On the contrary, Amazon could have become profitable on demand. They were just investing heavily into physical infrastructure back-to-back and if they ever wanted to slow down or stop, they'd be default alive.
Walmart back in the 80s investd heavily into physical infrastructure back to back - building new stores, their own trucks fleet. The first retailer with its own satellite communication system. There was never a question about Walmart going bankrupt.
I spent a couple of years in Amazon retail finance. The vendors were incentivized to sell as much as possible, under cost if that helps. Part of my job as financial analyst was to "convince" them it is a bad idea to sell under cost. Very hard and stressful work.
The part that I think may have some credence is that it may be a way to prop up huge profit margins that may persist for longer than they would otherwise in lieu of further commoditization and competition.
If NVIDIA has a stake in the company, they are less likely to do something like start brewing inference chips in house with the help of the foundry partner and a provider of vanilla chip designs. The company also gets a huge cash injection that is somewhat contingent on not doing that, and hey, they have a fresh pile of cash for cutting edge chips so whatever, right? My first thought was that these deals had more to do with that than anything else.
That aspect may end up being a bit illusory in the end. But then again, Nvidia has been proven to be quite skilled at building out and defending their ecosystem, sometimes through ruthless means, so maybe it persists (and I'm not sure that's a good thing). China certainly isn't a threat to throw a wrench in this situation... the entire US geopolitical complex will ensure that is the case.
The weird part is the _impression_ that $100B is being paid to a hardware company for chips, when in fact zero dollars are being paid to a hardware company and instead they are relying on public markets to spike the stock price based on potential future orders as if the demand is real.
The only reason this works is if we are the useful idiots buying up the stock.
There are sales taxes on each sale (GPU sales / Cloud VMs sales / etc...)
Meta point: governments want to tax each money transfer - when people are paid, when selling a real estate property, when selling stocks, when selling goods and services. Every time money changes hands governments want to take a piece. This is the easiest way for me to understand taxation in pretty much any jurisdiction.
"AI companies say these kinds of large, reciprocal, and sometimes overlapping deals are what's needed to meet growing demand for their tech. But as more and more money gets invested in the space these unorthodox business arrangements are also raising flags for industry analysts who are starting to wonder: Is the trillion-dollar AI market being propped up by the industry itself?"
It just we’ve seen the same thing for over four decades. I was getting stock certificates mailed to me from IBM in the 1980s when the savings and loan scandal rocked the market in 87. Then Enron. Then the internet bubble. Then the subprime meltdown. It just happens every decade or so when “too good to be true” runups happen.
Yeah I spoke to my wife a few min ago about these deals and other indicators of a bubble. We’re updating our 401k’s and the old college fund brokerage account in the morning and have agreed to not make any additional changes until Jan 2027. Going to sit out a year and see what happens.
Equities in the US are priced in dollars. If you think the dollar is going to lose value(because the government is led by people who have repeatedly told you this is their goal) then you expect stocks to rise commensurate with the devaluation of the dollar. I am not a bitcoiner or saying the USD is done for, but it seems clear that the dollar is weakened and will continue to weaken and this is a significant risk for folks like me with minimal exposure to stocks and precious metals.
Why would stocks raise with commensurate devaluation of the dollar? Stocks financial positions (assets / liabilities / cash flow) are most likely denominated in dollars.....
If companies' financial position would be denominated in something else (gold / Yen / Euros / CNY) this comment would make more sense to me....
Companies values are what they are. Equities are essentially their own currency. The denomination of their conversion price doesn't change their value. The listed price on the nyse or NASDAQ can be thought of the exchange price between an equity and USD.
If the USD loses half its value, the value of Microsoft does not change.
These sorts of outlooks are notoriously hard to predict.
Paul Tudor Jones (one of the most successful investors of our time) is of the mindset that if anything we are about to see equity appreciation far in excess of what we have seen for decades.
If that's the case, right when you're looking to rebalance your portfolio, you're going to be looking at a much more difficult decision, having already missed out on price appreciation that would be quite useful in padding downside risk...
Perhaps just step away from market cap weighted index funds as a long term adjustment. That's something that with a proper basic framework is advisable as a general portfolio management approach as well. (I have some SMLF - Smallcap Quality/Value, Berkshire, generic Global Equity exposure, not for outright returns, but to lower exposure to the concentrated passive market cap weighted indexes and diversify risk exposures).
Trend Following exposure is a great add as well, since it is negatively correlated with risk assets in many macro/market regimes.
Probably healthier to address your very real concern with modifying the long-term portfolio design rather than take a short-term market timing approach which is more of a negative expected value.
> These sorts of outlooks are notoriously hard to predict
The gut feeling you're about to lose your life savings isn't worth it for a lot of people, when I had most of my money invested it was always in the back of my mind. I cashed out and bought land/started building a multi generational house. All doubts and fear disappeared overnight, I'm for sure losing money compared to going all in on Nvidia or even btc, but I don't need any mental gymnastics to rationalize my choice.
> I cashed out and bought land/started building a multi generational house
interestingly, my wife and I have discussed this a lot too. We have 20 acres in SE Oklahoma that we got a really good deal on during the pandemic, the land is unplatted and the realtor didn't know there was water and power service already setup by the previous owner. We want to build a small cabin/house over the next couple years. It's like a retirement backup plan, "if everything goes to hell we can still go sit on our land in OK and watch the sunset..."
my 401k is setup in pretty high risk ( according to Vanguard ) index funds. Now that i'm older and looking to retire in the next 10-12 years, the investments need to be re-balanced anyway. As for the brokerage account, it's just a vanguard sp500 index fund. I'm not super exposed to the mag7 but they're such a huge component of the overall sp500 that when the bubble pops it's going to affect the sp500 quite a bit. I was actually looking at some of the smallcap Vanguard funds available to me because they seemed to fair ok in the .com bubble burst while the sp500 and especially the nasdaq were hit hard.
You are being mindful of some real risks and concerns.
That said, there's no reason to have to feel like you need to take a stab at it totally by yourself. Heck, that's what I'm doing in the day to day - putting together a few of these portfolios for people. There are certainly advisors and managers who know the academic side of this and can diversify away from concentration risks and such.
There are also many great tools in the relatively recent future that have opened up for common investors to diversify smartly. You can get pure trend following wrapped up in an ETF. You can get 100% equity 100% bond wrapped up in an etf at half of the size, so you can fit Gold and Trend in a portfolio with no hassle (both are low return but great from a diversification standpoint, so most retail investors want to lever up a bit to get back to the standard SP500 return/variance profile).
If you don't or can't find someone who can math it out for your individual case, it's also an option to just pull a few of the best proven individual ideas, such as "have some international diversification". ETFs are easy to grab here. Even 10-20% can pad out some of the US concentration and dollar exposure a bit. Smallcaps, like you said, are another way to shift away from the "size factor" dominated index weighted ETFs. The Nifty 50 bubble in the 60s was a major catastrophe, and history rhymes, so why not hedge a bit while still sticking to a proven investment framework? No financial advisor is going to look at an equity slice of 80% S&P, 10% international-ex-US, and 10% smallcap and say it's a stupid idea. Actually, they will probably say it's a smart improvement over 100% S&P.
When it comes to smallcaps, the caveat is that the Russel2000 is filled with fairly low quality companies. High debt, value traps, you've got it all. That's diversification, but probably not the type you are seeking. That's why smallcaps are a great target to add some factor based investing returns. This is another diversified source of return in and of itself (AQR as 150 billion under management here), but why not overlay some Quality and Value factors on top of the smallcaps? Now you have a sprinkle of factor investing edge in the portfolio alongside the smallcaps, while not drifting too far from the status quo since smallcaps are a small allocation. Not going to toss out specific investment ideas but there are ETFs/mutual funds for this! Just stick to low fees and a long track record. A question like "I want 10% in smallcaps, but I want a factor overlay for Quality and Value", is a question that would actually get money out of a good advisor by the way.
Here's a good, slightly wonky, lecture on some of the esoteric yet quite "in your face" risks within a complex dominated by passive and index weighted flows. It's not at all a misguided though to think there are some risks out there in the left tails. This was a lecture from 5 years ago: https://www.youtube.com/watch?v=x-rJciYZmi0&t=1s
If we end up with inflation rather than a market crash, this could be quite costly? Gold is up 50% this year, which hype is more correct than the other?
I am for a medium risk profile and I've already started diversifying. A lot of major "global" or "index" tracking funds are now majority comprised of holdings in Nvidia, Microsoft, Meta, Alphabet and also Tesla. I've seen many supposedly non-tech funds where about 50-75% of their portfolio are linked to this bubble. Sometimes the holdings aren't clear if the fund invests in other funds. As an individual investor I won't be able to react fast enough to a crash, so am being proactive with moving investments so that no more than half is exposed. Will obviously keep an eye on things to reduce that if it looks wise.
There is some circular financing going on, but AI accelerationists think this will be offset by demand, value, and adoption in businesses. Hence these deals are warranted for the incoming demand.
I think I saw Altman saying there's a global shortage of compute just now so this may address it. I'm not sure how much is actual user demand though and how much just investors wanting to pile into AI startups.
Upvote because can someone explain to someone as dense as me, whether or not this is likely to make some likely AI bubble worse? Is this just how industry allocates capital?
Part of what's concerning here is that the deals are conditional. OpenAI must meet XYZ conditions before cash/stock/etc is transferred, and the conditions are pretty hard to meet.
The money between OpenAI, Nvidia, Oracle, AMD is not circulating. There is no cashflow, only future commitments that may (and quite likely will) collapse. Yet the stock market & media react as if it's a sure thing. Even in the criticisms of these deals, the hype is affirmed.
This is the same problem as Enron's accounting, minus the fraud. (No need for fraudulent accounting when people simply don't read the fine print.)
The money is allocated by institutional investors. They buy the stock, the companies trade their more valuable stocks amongst themselves for various deals. If institutional investors stop investing then the flow of deals stops & the bubble pops. There is nowhere else the institutional investors could park their money other than tech so I don't think the bubble is going to pop any time soon. But infinite growth is obviously a logical & physical impossibility so eventually there will be a correction but whoever says they know exactly when that will happen is lying & they'd be better off buying lottery tickets to cash in on their ability to predict the future.
Personally think this makes a tech bubble a contagion for other parts of the financial industry - especially if institutional investors take the "easy" trade that everyone is doing and add leverage to it.
Now once these folks don't get 800B per year in revenue and the money runs out, all of the banks go as well. But don't worry - they'll get bailed out with our money...
Is it worse than the dot-com bubble? I remember everyone and their mom who knew HTML could get hired, and there were way more companies that went IPO during the dot-com bubble, like theglobe.com.
AI is a bubble, but is it worse than the dot-com bubble or the real estate bubble in 2008?
> Is it worse than the dot-com bubble? I remember everyone and their mom who knew HTML could get hired, and there were way more companies that went IPO during the dot-com bubble, like theglobe.com.
IDK, the number of companies and employees might be smaller, but the valuations and comp packages are so much bigger that I suspect there's more money in the bubble this time around, just a bit less spread out.
The dotcom bubble had investor capital going towards hiring tens of thousands of people. The AI bubble has capital going towards buying hardware, signing a handful of researchers on 9-figure deals and settling lawsuits for IP theft.
If the AI bubble bursts, there's a high chance the Everything Bubble bursts right after. And then USD goes gg.
I think part of the problem is investors are counting on infinite bailouts because AI is too big to fail. But that also causes USD erosion.
I have no clue what is the way out no matter who is in the driver's seat. And speaking of that, we had a decade of incompetent people in key positions at the Fed/Treasury, BoE, and CBE.
Stock buyback is effectively just a dividend with a different tax implication: reducing the number of shares in circulation raises the ownership stake of the remaining shares
> Stock buyback is effectively just a dividend with a different tax implication
Not just different, it specifically a lower tax rate assuming that the stockholder has held long enough to use the long term capital gains tax rate (which lower than the dividend tax rate).
Stock buyback isn't a total scam as it seems, but it does mean "we can't figure out any productive use case for this cash in advancing R&D or scaling our business anymore" which is still pretty worrying
It’s more like, “the executives with lots of shares can’t see how to make the company grow, so they’ll just use profits to pump up the share price for their gain”.
I deeply feel buybacks shouldn’t be illegal but treated shamefully.
Instead of using profits to build up long term savings or fund R&D, they basically choose to do as little as possible.
> Instead of using profits to build up long term savings or fund R&D, they basically choose to do as little as possible.
Isn't buying stock a form of long term savings? After all, they can always sell that stock when they want to "withdraw".
Sure, they may not get the same return as simply stashing it into a bank account, but it's also a statement of confidence in their future: "We are sure that our stock will outperform every other option there is for storing our money, because our long-term plans include extracting more profit from the market."
So far OpenAI hoovered - (minus) $7 Billion in first quarter 2025. $11.2 billion OPEX (compute, sales/marketing) in exchange for $4.3bn revenue. I wonder if revenue sharing agreement with Microsoft means Microsoft gets to pay for half of that :-)
Here is money printing scheme that looks to be at work:
Initial situation:
* Big corp M has X$ in cash where X is huge
Big Corp M invest X$ in AI startup O, with a provision that O needs to use most of the money to buy cloud infrastructure from M to power AI models.
End situation:
* Big corp M has X$ wort of shares in O, the value of which will rapidly grow
* Big corp M cloud division has ~X$ in extra revenue
The deal automatically turns X$ into ~2X$ in books.
Rinse and repeat with next round deals and next AI startups. The big corps are reporting increased cloud divisions revenue from AI spent, but it is their own investment money flowing back to cloud divisions.
> The deal automatically turns X$ into ~2X$ in books.
No it doesn't. The revenue it gets back is valued only at a fraction, because it's only worth its profit. Revenue != profit.
And your "the value of which will rapidly grow" is doing all the work here. That's not guaranteed. It might collapse as well.
It's not money printing at all. It's tying up cash long-term in exchange for a much smaller amount of profit short-term. Which is risky but entirely normal.
To be precise I wrote 'in books' which record also revenue, not just profits. Increasing cloud revenue is one of the things that drives big corps share price up, the missing bit is that this revenue comes from big corps themself. The growth investors mantra is that as long as revenue is increasing rapidly, they don't care much about today's profits, this is why so many unprofitable companies have crazy high valuation.
We are fully in the grifting stage of society. No one wants to work anymore and everyone wants to invest in imaginary outcomes (cryptocoins (most of which no one uses with little to no real use cases) and now AGI (science fiction)). Once the stock market crashes within an already low trust brittle society I imagine that there will be catastrophic outcomes. Maybe I am just a doomer who needs to touch grass but maybe not.
Welcome to how economies work. You only worry when you can’t see where the ends of the circle meet up. Ponzi schemes are notable specifically due primarily to the fact that they’re so directional and one-way. If you can’t pinpoint the escapement and see how it’s feeding into the winding of some other spring, walk away.
Welcome to how economies work. You only worry when you can see where the ends of the circle meet up. Ponzi schemes are notable specifically due primarily to the fact that they’re so directional and one-way.
> Two weeks ago, Nvidia Corp. agreed to invest as much as $100 billion in OpenAI to help the leading AI startup fund a data-center buildout so massive it could power a major city. OpenAI in turn committed to filling those sites with millions of Nvidia chips. The arrangement was promptly criticized for its “circular” nature.
That's not a "circular" deal, that's a... regular deal.
I genuinely don't understand the criticism here. Every business deal is -- I'll do something for you, you do something for me, and we'll both be better off. That's the free market.
Regardless of whether AI is a bubble, this is just business as usual.
A healthy two‑way exchange isn’t the issue, as you note, partners routinely buy from each other. The problem critics flagged here is that this particular arrangement starts to resemble “vendor financing” or a circular funding loop. Nvidia commits up to $100 billion to OpenAI. OpenAI then uses that money to buy Nvidia’s chips. Nvidia books the revenue and likely earns equity or upside in OpenAI. On paper, both sides look like they’re growing dramatically. But it’s the same capital being passed around, and it makes it harder to tell how much real, third‑party demand is driving those numbers.
Why people worry:
• Real demand vs. vendor financing: When the supplier is effectively fronting the cash for the customer’s purchases, it’s not the same as independent spending from end users. Investors and regulators want to see whether OpenAI could buy all those chips without Nvidia’s financing.
• Confusing financials: Nvidia appears to rack up huge sales, while OpenAI’s balance sheet swells with cutting-edge hardware financed by Nvidia. Untangling how much of each company’s “growth” is externally driven versus funded by the other side becomes tricky.
• Risk concentration: Nvidia’s fortunes would hinge on OpenAI delivering an multi-trillion-dollar scale business to justify the investment. If AI demand doesn’t materialize as hoped, Nvidia isn’t just missing chip sales, it’s directly exposed to OpenAI’s downside.
• Governance & optics: When vendor and customer roles blur, people question whether decisions are being made to satisfy true market demand or to juice financial metrics and valuations.
So it’s not that reciprocal deals are bad; it’s that circular, high-dollar deals can obscure what’s actually happening under the hood. That’s why folks are raising eyebrows.
It’s like Sam giving Pat a $20 “investment,” and Pat immediately spending that same $20 to buy Sam’s lunch vouchers. Sam says, “I sold $20 of lunches!” Pat says, “I’ve got $20 in lunches!” But really, it’s one $20 bill just doing a lap.
I'm generally of the same take as you: this reads much more to me like 2009-2013 Apple where they fronted cash to suppliers to do build outs and get to the 10x scale Apple needed ASAP. People looking at this as "proof" there's a "bubble" scare me.
How are these "circular deals" different than Boeing (and other heavy civilian equipment makers) providing financing on purchases? My point: Is this simply clutching pearls over the AI/LLM bubble, or is there some substance to this concern?
And another: How does it compare against the new breed of Bitcoin treasuries, like Strategy (MSTR.OQ)? As I understand, Strategy uses secondary offerings to continuously see more shares, then uses the capital to buy more Bitcoin.
This happens also with automotive companies. Peugeot has its own bank, which lends money to people buying peugeots. The big difference: is not circular. The bank is a bank, the OEM building cars build cars.
The bank makes money by lending and cashing interest.
The car maker by selling for profit.
The money goes from bank to OEM, and not the way around, no circle.
I sense a business opportunity here: betting on stocks! Maybe we can have a central market where people can exchange their money for bets. Maybe even sell their bet off before the due date.
We can call it a "stock market", or maybe even a "stock exchange". I expect I'll be able to get a patent on it if I prefix the name with "online", or suffix it with "on the internet".
The other day I was helping somebody choose some funds to invest in. The majority of "global" or "index" funds we looked into either directly or indirectly had the largest share of holdings in Nvidia, Microsoft, Meta, Alphabet, AMD. Often these were about 50-70% of their portfolio. The returns for the past year look great but the crash is going to be devastating and widespread because so many investments link back to this bubble.
There are plenty of non-US index funds, like EU, UK, Japan and others. There are also indices that track smaller companies rather than just the S&P500 or Nasdaq.
Or diversify in both directions - small US, and big and small international funds.
There are other companies than the magnificent 7. Other boring companies with boring dividends and solid earnings that have nothing to do with tech. Ulta. Pepsi. Chevron. But you sound pretty smart so I’m sure you had a good reason to invest in non us markets.
OpenAI is spending little to no money on the hardware and AMD can claim both investment expenditure and revenue. I made clear that I'm not sure the intricacies of how that or may not be tax evasion or fraud, though I dotn see how it could be neither.
I don't understand why they are calling these deals circular. OpenAI is buying Nvidia and AMD chips. Oracle is also buying Nvidia chips. OpenAI is buying datacenters from Oracle, which will be powered by the chips Oracle buys from Nvidia. This is one directional: hardware makers (Nvidia and AMD) sell either to datacenter makers (Oracle), or to AI firms (like OpenAI). That's it. No circular deals.
But "circular deals" has such a nice ring to it, that you hear it everywhere nowadays. People are just hungry for negative soundbites.
A stronger counterpoint to the circular deals suggestion is at the end of the article, which reads
> Michael Intrator, CoreWeave’s CEO, acknowledged the circular financing worries in a recent interview with Bloomberg News, but said the public concerns will dissipate as more businesses adopt AI.
> “When Microsoft comes to us to buy infrastructure to deliver to its clients who are consuming 365 or Copilot, I don't care what the narrative is about circular financing,” Intrator said. “They have end users that are consuming it.”
It's on you to decide then if end users are actually getting value / paying for AI products.
"Brutally honest and cheerfully sarcastic insights on Microsoft’s low-code and AI technology, business apps, and the business of software - written by Jukka Niiranen"
What credibility does this guy have? Sounds like someone that has an axe to grind.
In the end with most of these deals its the shareholders paying through market cap dilution. Given the current market structure (the big companies are tech companies) there's PLENTY of capital for OpenAI to fund their expansion.
They've discovered a cheat code IMO. Instead of using and raising money themselves, use their reputation/popularity and use their suppliers market caps (e.g. NVIDIA, AMD, etc). The deal makes sense as long as the value projected to be added (i.e. via efficiency gains, loss of jobs, changing society, etc) exceeds the capital dilution for the supplier; they use their equity but the leftover equity value increase makes up for it.
Given all the passive investing, and funds invested in the top tech companies this is a VERY large pool of capital. It however increases leverage if the value doesn't materialise.
Yes. OpenAI is also investing in AMD. This was discussed yesterday on HN, following some very good explanation by Matt Levine at Bloomberg. This is a way for one party to reduce some risk, by enjoying some upside in the equity of the counterparty.
But this is not circular. Circular would be if I sell you an apple worth $0.25 for $2.00, and then you sell it back to me for $2.00, or other similar amount, and I get to mark all the apples in my inventory at $2.00 and show a huge profit (on paper). One can create variations of this blatant deal. Like I sell you some rubber for 10 times the market price, you make a balloon and then I buy the balloon for 10 times the market price. I may not have other balloons in my inventory, but plenty of rubber, and I show some nice profit. One can imagine other, fancier deals.
But in the case of AMD and NVidia, and OpenAI and Oracle, the direction is clear. OpenAI has a clear need for compute. They can buy it directly from NVidia and AMD, or indirectly from Oracle. They can buy it with hard cash (of which they don't have that much), or with their own equity, or some form of deal that offers the seller an upside in OpenAI's equity.
But there is not back and forth buying of the same item, or of rubber/balloons. All the deals seem legit. Is it possible that all the future compute will not be needed, because the AI craze will fizzle. It is, lots of things are possible. But that's general business risk.
> But this is not circular. Circular would be if I sell you an apple worth $0.25 for $2.00
When you draw a circle on a piece of paper do you put your pencil to the paper, start drawing a curve, stop and write the word FRAUD, and then complete the curve? Or do you just draw a circle?
I’m assuming that people are saying “circular” in that it looks like the money goes in a circle. (For one example) Nvidia invests in Lambda, Lambda buys GPUs from Nvidia, Nvidia leases GPUs back from Lambda, Lambda uses the revenue from leasing the GPUs to Nvidia to raise debt to buy more GPUs from Nvidia…
This is how financing works. When you buy a house, you get a mortgage from a bank. It is unusual to get a mortgage from a seller. It would feel a bit circular, right? But that is exactly what happens most of the time when people buy a car. They get a loan from the same company that sells them the car. Is that circular?
When you replace people with companies, the financing can become much more complex. The example you provided with Nvidia and Lambda seems quite reasonable to me. Here's an example that happens every day in the world of housing: banks lend money to house buyers. Then they package the mortgages and sell the resulting mortgage back securities. Then they take the money from the proceeds, and give more loans, and package them and create more mortgage back securities. Seems circular, right? But that's just how business is done. There is no Ponzi aspect to this, or fraud, or smoke and mirrors. It's just every day business. Nobody labels that as being circular.
When people buy a car they sometimes get a loan from the auto manufacturer’s financial services arm. What they don’t usually get are warrants struck at a penny for 10% of the manufacturer.
You sound like you know what you’re talking about. I only think I know what I’m talking about. Help me understand: What am I missing in the OpenAI / AMD deal that makes it non circular?
OpenAI: We need lots of GPUs, you make GPUs, but your GPUs are not quite equal to Nvidia's GPUs. And we are a growth startup, we don't have a lot of cash, can you give us a hefty discount on those GPUs?
AMD: We need cash too, you know? We'd love to sell you those thousands and thousands of high end GPUs, that would cover some of our R&D, and we could one day match NVidia. But we don't swim in cash either. We can't give you the discount.
OpenAI: What can you give us? You must be able to throw in something there. Otherwise, honestly, we can't make the deal.
AMD: What if we give you some equity? And if our deal goes well, and our GPUs start being viable alternatives to Nvidia's, maybe we'll be able to get close to Nvidia's market cap and even surpass them, just like we did with Intel.
OpenAI: Brilliant. We love it.
AMD: Yes, but that equity will be contingent on how the deal goes.
> You sound like you know what you’re talking about.
Honestly, I'm not an expert either, but I've run a company, and I can all but guarantee that credit_guy above really does not know what he is talking about.
I've replied else-thread, in detail, on exactly why his analogy to mortgage and car loans are incorrect.
His main point that "these deals cannot be circular because mortgage and car finance are not circular" is incorrect. The mortgage and car finance deals are not analogous to the Nvidia/OpenAI or AMD/OpenAI deals.
The AI startup valuation largely is. I feel it quickly becomes circular because people make projections purely on other projections since the world is too impatient to wait and find out.
A single hamburger store is never going to be projected to have a billion dollars of revenue because people understand the total addressable market. Doesn’t matter how good the burger is.
The AI stuff is too new that people don’t have a firm grasp on the costs and profit opportunities. They don’t really even know how to define the TAM. Too many unknowns. Entire classes of labor could be replaced by AI —- or perhaps not.
With little grounding, it quickly becomes a circle of hype.
I don’t know what to tell you guy, but when people see money moving in a circle in a deal there is a good chance that “circular deal” might pop into their heads. Because it’s a deal that is shaped like a circle.
> Is that circular?
Doesn’t really seem so because at the end of the day the money goes from me to them. I don’t get my money back, I get a car in exchange for my money.
Also this deal didn’t begin with the manufacturer purchasing shares of me before offering me debt to buy a car from them.
>But that is exactly what happens most of the time when people buy a car.
Your car manufacturer leases your car back from you? And you use that revenue to raise debt to buy more cars from them? What manufacturer are you doing this with? What do you end up driving?
> Your car manufacturer leases your car back from you? And you use that revenue to raise debt to buy more cars from them?
Honestly, if this was possible people would be doing it (not that they are not - fleet services and rental fleet services do some pretty funky accounting sometimes, so I wouldn't be surprised at all if this sort of thing happened).
If this was possible, I'd be doing it.
> What manufacturer are you doing this with?
Good question. The minute an answer gets posted I'm going to have a really good side-hustle.
> What do you end up driving?
I presume, at the point that this is being done, there is no actual car involved, so nobody involved will be driving because there is no car to drive.
> Good question. The minute an answer gets posted I'm going to have a really good side-hustle.
Yeah “everybody buys cars through the infinite car glitch” sounds like the sort of thing that would be part of an enormously long answer to “should I try WoW for the first time in 2025”
"Do not charge your brother interest on money, food, or any other type of loan. You may charge a foreigner interest, but not your brother." - Deuteronomy 23:19-20
> But that is exactly what happens most of the time when people buy a car. They get a loan from the same company that sells them the car.
No, they don't!
Serious "Citation Needed" here. They get a loan from a financial services company, that is a separate company from the automaker and/or dealership.
The certification and requirements for trading as a credit provider will not be met by neither the auto manufacturer nor the dealership.
> Here's an example that happens every day in the world of housing: banks lend money to house buyers. Then they package the mortgages and sell the resulting mortgage back securities. Then they take the money from the proceeds, and give more loans, and package them and create more mortgage back securities. Seems circular, right?
No, it doesn't! They make out new loans, sure, but they aren't loaning you specifically the proceeds from the sale of your specific mortgage-backed security!
If you happen to get a new loan from the sale of the MBS, it is impossible that only you get that loan, from the sale of an MBS that had only your existing loan.
Seriously, there's laws and regulations around this, and from what you say, with all due respect, it seems that you are unaware of these regulations.[1]
The only reason that these actually-circular deals can go on right now is because OpenAI (and other providers doing similar circular deals) are not publicly traded, and thus there are fewer regulations and even fewer enforcement of what little regulations there are for unlisted companies.
=================
[1] Why is your handle "credit_guy"? You don't appear to be familiar with the fact that credit providers are heavily regulated in all jurisdictions that we are talking about. I mean, you don't even need to know the specific regulations and certifications necessary, you just need to know things like a dealer cannot be a credit provider too.
> But there is not back and forth buying of the same item, or of rubber/balloons. All the deals seem legit.
Some, certainly, but the vendor-financing deals certainly look circular to the casual observer. Microsoft invests $X into OpenAI for a 51% (or whatever) stake, and that investment then goes straight back to Microsoft to pay for compute credits.
Or Nvidia invests[1] $100m into OpenAI, which OpenAI then turns around and pays back to Nvidia for compute.
The majority of the deals making the news are structured like this; maybe technically those aren't actual ducks[2], but they sure look, walk and talk like ducks.
Similarly structured deals are with Oracle. And Coreweave. And everyone.
It may not be a "circular" deal, but what do you expect people to call it when a company makes a deal to receive cash (not credit, but actual cash) from a vendor, and spends that cash with that vendor?
==========================
[1] I use this word loosely here - the investment is a commitment of 10x $10m tranches.
I agree with your general point: there is a tendency to group not least these deals together without distinction as confirmation of a pre-existing belief that the AI bubble is soon to collapse. The latter certainly might be true, but we should be careful about too quickly deciding what evidence supports it. I am inclined to strictly separate:
(a) the NVIDIA deal as a vendor subsidizing one of their largest clients, which perhaps signals market weakness;
from (b) the AMD deal as a non-market-leading vendor trying to entice a new enterprise client, who will only be able to use AMD's chips for serious training after significant (and risky) collaboration to improve their product offering.
This distinction is important to me because I see more concrete evidence for a possible material drop in NVIDIA's business short- to medium-term than a collapse of the sector as a whole. The chips act clearly shut them out unexpectedly from their second largest market and now it seems likely that Chinese chips will be competitive for training sooner than expected. Indeed, if Chinese AI firms are suddenly able to obtain even a roughly approximate product at considerably less cost, OpenAI will suddenly find themselves at a cost if not compute disadvantage to their Eastern competitors. It isn't a surprise, then, that OpenAI is now looking to reduce their present vendor lock-in with NVIDIA.
Overall, it's not great for the latter if they lose access to their second largest market, suddenly have viable competitors in their home market, and either lose some of a major client's business or have to significantly reduce pricing to retain it.
https://archive.is/dvKkB
This reminds me of around 2002 when I wrote an article looking at how all the web behemoths at the time were claiming profitability through ad sales, but actually the vast majority of ads were from one web behemoth advertising on an other's site and vice versa.
Same as capacity swaps in telecom of that era
This is How Larry Elisson beat Musk for a short while although Oracle was struggling.' The deals among Oracle, Nvidia, and OpenAI have raised concerns about their circular and potentially risky nature. Oracle is reportedly spending tens of billions on Nvidia's advanced chips, Nvidia plans to invest up to $100 billion in OpenAI, and OpenAI uses Oracle's cloud infrastructure through Microsoft's Azure, creating a closed loop of financing and business. Some analysts warn this creates a reflexive loop or "dangerous bubble" where valuations may be inflated artificially by the circular flow of capital and resources, reminiscent of past booms that ended in sharp market corrections.
If this means that the fall of OpenAI will cause Oracle and Nvidia to crash and downsize, bubbles aren't completely bad.
You can't discount the impact this will have on global stock markets and what that may do to both individuals and, more importantly, pension funds, as a large swath of people are retiring
Damage is unavoidable, we can only hope that it happens to someone deserving. Innocent individuals and pension funds still have time to retreat; if they don't it will be their fault.
We are moving towards gerontocracy - if pension funds will have large losses it’s very likely that young, working age people will be taxed extra heavily to keep the QOL of pensioners.
That would likely lead to a revolution: Millennials are 30-45 and they’re not in a good place; neither is Gen Z.
We’re already seeing revolutions elsewhere — and it’s likely that trying to loot them further by generations who sold out the nation will simply lead to social collapse.
Have you ever battled an 87-year-old wearing mechanized battle armour? They're crazed, hopped on speed, eyes goggling in their sockets
A pack of 3 oldies burst through our perimeter one winter night... the screaming woke me up. Outside my tent the forest was lit up red by our laser blasts, trying desparately to take them out.
We thought that a revolution would be a good idea, but an upside-down population pyramid is a hell of a thing when you're on the bottom.
Perhaps most of the impact could be burdened on NVIDIA investors, employees, leadership, or OpenAI investors, employees, leadership, etc., etc.
Yes but this is by design. Wealthy people want bubbles, when they pop, you can gobble up all the value at all time lows. Then you hold until you don't feel like it anymore as the market goes back to growing. You see this as how private equity has bought up real estate across the USA to turn into rentals after 2008, for example.
Aren’t extremely wealthy people that wealthy due to the valuation of their stock? IIRC generally the higher the networth, the higher share is kept in stocks
Wealthy people are extremely well diversified. Their main risk exposure is to variables that effect the entire market - e.g interest rates.
And they don't spend money, they take debt against their existing assets to fund projects and investments. So long as they can service the loans across economic downturns, they don't particularly have to feel the effects of a recession, outside of the mentioned opportunities to buy the market at a discount.
I suspect $$ is just a number for them. Being able to control more resources is the ultimate game. You gotta have zillions of $$ to join the tournament, though.
Person A has a net worth of 2B Person A has a loan at 500M backed by their holdings Stocks drop 50%, Net worth is now 1B Person A buys $500m of stocks Market Recovers 100% Person A now has 2B original holdings and 1B gains, $500m owned = 2.5B
Very simple example, and not the only way to do it - but people need to remember net worth being 500B is not 500B in the bank, and at some point the number doesnt matter
They can move their wealth from stocks to gold, for example. Look at the price of gold.
More importantly you keep the portfolio semi-balanced.
Just using Google / Gold as a comparison [1].
Assume you have 100 units of each.
In late 2021, Googs gone up ~100% so you have to rebalance because you have $200 in Goog and $92 in Gold. So lets say you rebalance to 80 Goog (160$) and 144 Gold ($130).
In late 2022, Googs gone down ~40% so you have to rebalance because you have $96 in Goog and $141 in Gold. So lets say you rebalance to 100 Goog ($120) and 118 Gold ($112).
So over the course of 2 years Goog has gone up 20% and Golds gown down 5% but your investments are overall up 16%. Obviously a 100% Goog investment is higher but with more risk.
If you didn't do any rebalancing then you have a gain of 7.5% (100*1.2 + 100*0.95 = 215)
[1]: https://www.google.com/finance/beta/quote/PHYS:TSE?compariso...
There was an article here recently about alpha school that mentioned this is how of of its backers made a lot in the aftermath of the dot com bubble - picking up companies with products that happened to implode in the bust. He's not the only one and that's not the only bubble that had that happen
> If this means that the fall of OpenAI will cause Oracle and Nvidia to crash and downsize
Only if those are the only actual real customers. It's not. The pie is a lot bigger. And with the current hype some other AI company will just take over and the bubble will continue.
Where does that leave AMD, considering OAI are taking a 10% stake?
A shareholder going bust is a no-op. However, the loss of large customer, is concerning, but less so, relative to the market souring on AI investments as a whole, which would be not great for AMD, but disastrous for Nvidia.
They'd just pitch for using the same GPUs to run the digital dollar ledger, and begin the cycle again, maybe.
Sure who cares if we wipe out 15% of everyone’s retirement right as long as it’s got some schadenfreude involved.
I doubt, or at least hope, that no sort of managed fund for retirement (read: low risk) is going 15% on AI stuff. Regardless of what one thinks might happen, it has to be accepted at this is an obviously very high risk bet, because this all is screaming bubble even more than the era of 'zomg i flipped this house i financed on an insta-approved loan, installed granite counter tops, and walked away with $20k profit in 4 days.'
But probably they will invest in the sp500, and tech companies make a big portion of the sp 500 and these tech companies also happen to be big in AI
Failing to diversify is the fund managers' problem. Pension funds shouldn't have all their investment in a single countries' stock market (doesn't matter that most, if not all, of those companies operate worldwide, they are still from the US), not even in equity markets. A wiser investment manager (or individual investor) would put a chunk in different kind of equities, and diversify into fixed rate, precious metals, perhaps real estate... Not bet it all to the volatile S&P 500. "Past performance does not guarantee future results.".
Who's "we"? Most of us have zero say when the backroom circular deals are made.
If it’s a bubble, then the long term value people are now counting on to retire with in the future is not actually real though, right? Like a bubble popping doesn’t destroy real value that used to exist, it exposes fictional value that never really existed.
It reallocates real value to the capitalist class. Most of us don't babysit stocks 8h a day and will be holding the bag.
What’s the mechanism you have in mind for that happening via bubble? They sell early before the crash? I’m curious if that actually happens, since the mass psychological nature of market crashes seems to make them difficult to predict reliably, even for people with a lot of resources.
Inflation on the other hand definitely does reallocate real value to whoever already owns a lot of assets, and I believe inflation often goes along with bubbles, since credit expansion is what usually kicks off the bubble in the first place.
Is it fair to say OpenAI is in a sense “washing” the money passing between Nvidia and Oracle? And instead of taking a cut in the traditional money laundering they are enjoying massive valuation gains?
When the fed investigates this does it matter if one of the 3 companies is not a publicly traded company?
I guess you could see it like that but I don’t think it’s “wash trading” in the sense that they’re coordinating it. I don’t really know what the fed would investigate here.
What you're missing is that real value can be created in these exchanges - Nvidia really makes chips for example.
The issue is that it's an industry of investment which exists solely to power more investment in AI - the entire chain is still assuming that someone will eventually pay for this.
At the end of the day all that money leaks out to employees and suppliers...but no one those people transact with may have any interest in buying what was produced.
That's only really value if the chips are useful and if there are people buying the chips for something they want to do with them.
It's entirely based on the perception that LLM training & inference is here to stay at ever growing scales when the shortcomings of Artificial Dreaming are increasingly scrutinized. Not all businesses want to end up paying refunds to their clients like Deloitte [1] because the LLM hallucinated crap into their reports (and they failed to correct it).
[1] https://www.theguardian.com/australia-news/2025/oct/06/deloi...
Deloitte just bought like 400k Claude enterprise licenses. Bad example.
You could argue about the causality here.
> It's entirely based on the perception that LLM training & inference is here to stay at ever growing scales when the shortcomings of Artificial Dreaming are increasingly scrutinized. Not all businesses want to end up paying refunds to their clients like Deloitte [1] because the LLM hallucinated crap into their reports (and they failed to correct it).
This assumes Deloitte didn't make more $$$ from the deal by "outsourcing" it to AI than not. It was a partial refund.
You haven’t brought in the side costs into your analysis
> You haven’t brought in the side costs into your analysis
And you haven't considered that Deloitte might get this money some other way anyway. It's been budgeted and needs to be spent by the government department for this financial year. They'll find another project to hand out.
lol you and I are not talking about the same stuff mate
Leaks out? Don’t you mean “trickles down”?
Oracle most certainly is not struggling. Even putting aside the NVidia deal, their cloud business is thriving.
They give you 4 nodes with 1 vcpu and 6GB RAM each for free forever by the way. Most other stuff has a free tier too.
I used it to learn terraform and have a kubernetes cluster running.
I know it's always trendy to bash oracle and simp for AWS, but it is THE best option for learning and hobbyists.
On a sofware salary, would rather pay the 20 something to a provider like hetzner instead of giving my personal info to oracle
That's very cool. I was going to get a hetzner for some hobby projects.
Can you get that without adding a payment mechanism? For some reason, I'd be more cautious about putting my card into Oracle's offering than anyone else.
> Can you get that without adding a payment mechanism?
No!
There are actually two free tiers.
The limited free tier you get without adding a CC and then this advanced "always free" tier I am using, where you have to use and verify a CC and have to pay big bucks if you mess up.
limited free tier: 300$ to be spent in 30 days
advanced always free tier: https://docs.oracle.com/en-us/iaas/Content/FreeTier/freetier...
So it's free... Until is not. My take on this would be that if Oracle makes any change to that "free" tier and you do not take proper action you might be automatically charged hefty amounts from your credit card. No, thanks. It's way way safer to spend few bucks a month in other VPS options than get a big surprise on your credit card.
Privacy.com has always been a working general solution to this, for me. Disposable CC aliases with spending caps.
Privacy.com won't provide lawyers to you when Oracle sends letters of demand to you for unpaid invoices.
I know Oracle are notorious for scummy license enforcement, but in practice their Always Free tier is superior to Amazon's precisely because they will cut off your service rather than bill you arbitrary (huge) amounts.
The big limitation is the 10mbit egress bandwidth, but you can work around that with (free) Cloudflare.
Messing up with Oracle is a bit different from messing up with Google. I'll go with Hetzner which I love anyway. :)
You can also get all of that in a single node, which makes for a fairly beefy build server :)
Uh, you can get that level of compute locally easy peasy.
Learning terraform isn't hard, it's what you stitch together with terraform that is hard.
I think the point here is that free beats easy peasy, especially when learning a new skill, where your easy peasy isn't their easy peasy. And getting something up and running on Oracle Cloud with Terraform is nowhere near easy peasy for someone who never did anything with Terraform ;)
'Back in my day' we learned with VMs for free! The computer you have works without internet, for instance
> while although Oracle was struggling
In terms of actual earnings not sure Tesla is doing better.
It's worth noting Elisson is 2 years older than Trump.
What's he going to do with all that money, and what does he care for the risk it's bad or shady?
Worst case, he got to be #1 for a bit for a few dozen billion, best case he's hoping AGI will extend his life before he croaks.
I miss the era when people like Dale Carnegie would sponsor some public buildings, instead of ruining the internet and the natural environment.
You may be thinking of Andrew Carnegie. Who was a steel industry tycoon, probably the #2 most polluting industry today. Back then - probably #1.
The Sacklers donated a lot of money to cultural institutions while they poisoned society with opiates.
Doubt he believes that. Getting AGI (whatever that actually means) means we are still far away from reversing ageing.
Even smart people belive a lot of untrue things, and his domain of knowledge isn't likely to include much more biology than my GCSE grade C and a handfull of brilliant.org micro-lessons; it's quite possible, though I would not put specific odds on it, that he's seen e.g. the sudden change in the rate of progress in protein folding, where before it was ~ one PhD thesis per protein and then suddenly AlphaFold did all of them, and extrapolated from that to everything else like an over-enthusiastic PopSci reporter.
(Interestingly, some of the world's dictators do seem to have an interest in the current state of the art in prolonging life. For example Xi and Putin chatted about organ replacement https://www.bbc.com/news/articles/cr70rvrd41ko)
Yea, but if you goal is extending life then investing in science is probably a better bet than GPU:s
Ah yes, my favourite philosophical dilemma, The Organs of Theseus, and whether or not a sufficiently organ-replaced person is absolved of their previous actions by virtue of not necessarily being the same person.
As a more serious point related to this though, I was under the impression that organ replacement didn't address issues with telomere length?
[flagged]
> What's he going to do with all that money, and what does he care for the risk it's bad or shady?
The thing that unites Ellison, Trump, Musk, Thiel and a fair few of the other politically active billionaires is the obsession with "legacy" - they want to leave their mark in the history books, figures which will likely be remembered and taught in schools in thousands of years similar to Roman emperors.
Musk is the most obvious with his obsession of settling (and eventually dying) on Mars, Trump is dreaming of getting a Nobel Peace Prize (if only to not let Obama be the only US President who got one), and the rest is hunting for the "invented AGI" crown.
Obama was the fourth U.S. president to be awarded the Nobel Peace Prize, after Theodore Roosevelt (1906) and Woodrow Wilson (1919)—both of whom received the award during their terms—and Jimmy Carter (2002), who received the award 21 years after leaving office. - wikipedia
Thanks for the correction!
Thiel wants his legacy to be bringing about the Antichrist…
Give it all to the IDF.
What does Ellisons personal wealth have to do with this? The concern is that the circular pattern of shifting money between these companies is artificially inflating the stock market to heights that will crash very very badly when this bubble finally pops.
The market is due a correction irrespective of bubble mania.
The implied equity risky premium is getting pretty low - a sign of greed.
Also a pg essay from 2010: https://www.paulgraham.com/yahoo.html
>By 1998, Yahoo was the beneficiary of a de facto Ponzi scheme. Investors were excited about the Internet. One reason they were excited was Yahoo's revenue growth. So they invested in new Internet startups. The startups then used the money to buy ads on Yahoo to get traffic. Which caused yet more revenue growth for Yahoo, and further convinced investors the Internet was worth investing in. When I realized this one day, sitting in my cubicle, I jumped up like Archimedes in his bathtub, except instead of "Eureka!" I was shouting "Sell!"
That's not how Ponzi schemes work. Yahoo had a defacto _monopoly_ and the market had bad discovery leading to bad price information. There was no point at which the internet was /not/ worth investing in and everyone who had experience with it knew that.
The real problem seemed to be that you can only put so much money into pets.com before it becomes stupid. You had more short term investment capital than could be _effectively_ spent at the time. The long term players, as usual, avoided the Archimedian idealism, and were heavily rewarded anyways.
pg has startup brains.
Same issue with LLMs - there probably is real value there, but investors have rammed 10 years' worth of investment into the market in 10 months.
That means: 1) Late investors end up taking a bath because it will actually take 10x longer to get a return than they thought. 2) Investment becomes inefficient - there are lot of GPUs being bought in 2025 that should have been bought in 2030. By the time 2030 comes along, those 2025 GPUs will be outdated, so the value they will provide will be less than if they had been purchased at the correct time
pg has completely clairvoyant vision, if only about historical events
>There was no point at which the internet was /not/ worth investing in and everyone who had experience with it knew that.
I don't get this. Of course there was. Specifically, March 10, 2000.
To declare that something is not worth investing in doesn't imply that it's useless. (and actually, swing investors will argue something can be worth investing in even though it is in fact useless, e.g. web3 and most crypto).
I'd take some Amazon shares at the March 10, 2000 price (inflation adjusted). Merci.
Why is it so popular to "akshually, Ponzi schemes are different!" these days?
Words have meaning in context, and calling something a "Ponzi scheme" these days means any pyramid-like system that only works as long as new investors come to the table, because that new investor money is the sole source of gains for earlier investors. But once you run out of new investors, which is inevitable, the whole thing collapses. What pg described was exactly a Ponzi scheme in that sense, even if it wasn't a deliberate scam. And that's very different from other types of business ventures that eventually fail because of the more "normal" reason that they just don't gain traction.
> Words have meaning in context
And you do recognize the context of those words was 15 years ago at this point?
> was exactly a Ponzi scheme in that sense, even if it wasn't a deliberate scam
There are better terms of art to describe the scenario. If we lacked for those words I might grant you this point, but since we don't, I find the description lacking in precision if not completely faulty.
> the more "normal" reason that they just don't gain traction.
Is that actually the "normal reason" most businesses fail? I'm not sure it is.
Agree, people became extremely pedantic with words and it's incorrect to assume they only mean what is described in a dictionary, as if we are in a court.
Language is fluid, and many of words got their very known & popular meanings later, when people used for things that it didn't exactly mean what it was initially intended to, and it got popular.
People are ever less empathetic and sociable, this being the reason for such comments.
People become pedantic when it's the only way to defend themselves and their friends or to argue against someone.
In this case dotcom investing was so stupid that it should have been recognized as stupid at the time, without hindsight. For someone who lost money, gave bad advice etc. insisting that they didn't fall for a Ponzi scheme means representing themselves as less terrible investors, regardless of facts.
This time it's different?
Do you happen to have a copy of that article? I’d love to read it.
No, but I could give you several articles that link to that article while talking about how great it is.
But you can't fake profits that way (without cooking the books); you can only fake revenue. That's why profits matter.
Though I guess with high velocity of money circulating in tight loops, everyone involved can feel rich. 1 million dollars changing hands once per day can potentially mint 365 'millionaires' during the course of a year. If they just pass it back and forth among themselves to buy each other's stocks and products and don't let that money escape their network to actually pay for something useful, they can all be millionaires... On paper.
Not to mention how relatively small amounts of money, moving at high velocity, can inflate asset prices... On Alice's birthday, Bob can buy 100 shares of Alice's shell company at a price of $10 per share. If Alice owns 1 million shares of the company (which she founded), it means that Alice's net worth is at least $10 million... It cost Bob only $1000 to give Alice a net worth of $10 million. Now imagine that same $1000 moving in and out of that company's stock, traded 1000 times per day between various people. With just $1000 circulating back-and-forth at high frequency (A.K.A. high-frequency trading), you can generate a trade volume of $1 million per day... So it all seems reasonable; a company with a $10 million valuation with $1 million daily trade volume... Nothing suspicious. So you can basically start a shell company and fake everything; from the market cap, to the trade volume; using only a relatively small amount of money. This is what they were doing in the early days of crypto.
The same money can hop around in circles between an insane number of people when it's just 'revenue' because revenue is not taxed (after expenses). Only profits are taxed... But the same money, as profit, can barely hop between 6 people before it's taxed down to just 10% of its original number. High-velocity revenue can severely distort people's perceptions of the market, especially in the era of media filter bubbles.
You can fake profits. Company A hands $1m to company B to buy equity in it. Company B hands the $1m to A to run ads on A.
Under GAAP (normal accounting) A has a profit.
In reality are they profiting? Who knows - it depends how A does as a business.
Damn. Didn't think of that. Seems I have a very naive definition of 'profit'.
It's crazy how all these small subtleties in definitions of various terms can create something so abominable in the aggregate. Makes one want to reach for a tinfoil hat. Not sure I can even trust the company which makes the tinfoil at this point.
Dunno about naive but it's more complicated than you might think.
In the very short term yeah, but in the long term GAAP still sniffs this out. Eventually Company A will need to use some form of mark to market valuation to assign a value to that $1M they invested. If Company B is a dumpster fire then eventually the $1M investment will be worth much less (or zero) and balance is restored to the world.
You see companies hand waving this sort of thing away on earnings calls when they have a really bad quarter but the CEO is blowing smoke at shareholders saying “On non-GAAP this was an amazing corner.” Which is code speak for “that terrible decision I made a few quarters ago is coming home to roost and our lawyers have said I need to reflect that on our books. If you ignore where we screwed up badly, our numbers look good!”
> it means that Alice's net worth is at least $10 million...
Some people might interpret or represent the aforementioned transaction as justification for thinking Alice's net worth is at least $10M, but it is not an objective measure of Alice's purchasing power (as it would be in the case that Alice had $10M cash).
>The same money can hop around in circles between an insane number of people when it's just 'revenue' because revenue is not taxed (after expenses). Only profits are taxed
On the US federal level. Not sure about other countries, but quite a few other governments in the US do tax revenue.
https://taxfoundation.org/data/all/state/state-corporate-inc...
>Nevada, Ohio, Texas, and Washington impose gross receipts taxes instead of corporate income taxes. Delaware, Oregon, and Tennessee impose gross receipts taxes in addition to their corporate income taxes. Some localities in Pennsylvania, Virginia, and West Virginia likewise impose gross receipts taxes, which are generally understood to be more economically harmful than corporate income taxes.
See also:
https://en.wikipedia.org/wiki/Gross_receipts_tax
All clearly designed to cash in on the speculative value being given to AI companies.
Strategically I thought OpenAI's deal of getting 10% of AMD for driving their stock price to $600 was a pretty clever way of creating $97 Billion from nothing - effectively paying for the GPUs they'd purchase.
At the same time I would've thought this insider pump and cash-in strategy would be somehow illegal, but I guess anything goes with this administration.
These sorts of shenanigans happen under whatever administration happens to be in office at the end of a long business cycle and near the frothy peak of a hype cycle. And that frothy peak can be quite tall and last for years.
it'd be illegal if they bought on the public market... but they bought directly from AMD, so nothing to see here, move along. music is still playing.
Just getting options and trying to get the company to do well isn't illegal or discouraged. Price manipulation can be illegal but I don't see any evidence of that so far.
This is going to end badly — and soon. It’s ironic that catastrophic forgetting (i.e., the inability to perform continuous learning) and hallucinations (i.e., the failure to recognize when a prediction is unfounded) won’t be the causes of the crash, but rather greed and stupidity.
Has greed and stupidity ever not been the underlying cause of a financial bubble?
The 08 financial crisis and mortgage backed CDOs?
There is a beautiful scene from "The Wolf of Wall Street", in which Matthew McConaughey outshines Leonardo DiCaprio.
https://www.youtube.com/watch?v=Y4iBdIq0aaY
I think it serves as a good commentary.
I finally got around to watching that a year or two ago. That scene is among several that are just remarkable bits of work, but somehow the whole film itself felt like a slog.
I feel like that's the problem generally with late career Scorsese. The last of his films that felt consistently interesting and free of flab was, to me, Casino, though he got close for me with The Aviator and The Departed.
Agreed on TWoWS. Kind of loved every scene in the movie, but actually strongly dislike the movie in full. I also think that despite what seems like an attempt to critique the destructive lifestyle, it does the opposite and glorifies it. Mostly not a fan of his later career work, but I do really like Killers of the Flower Moon and Shutter Island.
I'll be honest and say I never saw Shutter Island in part because I read the book (at one point I was a Lehane fan) and saw the twist coming miles and miles away.
I haven't seen KOTFM b/c it's past my Scorsese Endpoint. ;)
I can smell the alcohol in that guy's breath over youtube...
Amazing actor.
And plop! goes the bubble.
Anecdotally, many people at OpenAI, Nvidia, Oracle, etc., sincerely believe their company's own hype.
They remind me of the story about the oil prospector in Warren Buffett's year-end 1985 letter to Berkshire Hathaway shareholders:
> An oil prospector, moving to his heavenly reward, was met by St. Peter with bad news. “You’re qualified for residence”, said St. Peter, “but, as you can see, the compound reserved for oil men is packed. There’s no way to squeeze you in.” After thinking a moment, the prospector asked if he might say just four words to the present occupants. That seemed harmless to St. Peter, so the prospector cupped his hands and yelled, “Oil discovered in hell.” Immediately the gate to the compound opened and all of the oil men marched out to head for the nether regions. Impressed, St. Peter invited the prospector to move in and make himself comfortable. The prospector paused. “No,” he said, “I think I’ll go along with the rest of the boys. There might be some truth to that rumor after all.”
Source: https://berkshirehathaway.com/letters/1985.html
Both Jensen Huang and Sam Altman seem to believe.
Here's them 2 weeks ago being interviewed https://www.youtube.com/watch?v=jVBzsg3yCK4&t=40s
Also Brockman:
>And so, you really want every person to be able to have their own dedicated GPU, right? So, you're talking order of 10 billion GPUs we're going to need. This deal we're talking about, it's for millions of GPUs. Like, we're still three orders of magnitude off of where we need to be. So we're doing our best to provide compute availability, but we're heading to this world where the whole economy is powered by compute... (15 min in)
"this world where the whole economy is powered by compute"
Is actually laughable.
Rewatching the video it strikes me that there is no talk of how the compute will be paid for. Altman is saying there is so much demand for compute but a lot of that is because he's setting the price for users to zero in spite of the service costing a lot to provide.
I guess they figure once they have the users they'll monetize somehow but that bit's kind of iffy.
Is there? Is there actual organic demand for compute or is it all just a mirage created by the companies themselves?
I agree, though it's not out of the question that there could one day be a single country whose economy relies entirely on data centers, similarly to petrostates. Maybe we'll call them datastates, or compustates.
We have that today - Sealandia! Also where does today's Taiwan fit in that spectrum? Is it a "compustate"?
I agree with the idea but it is no certainty.
If their compensation includes stock options, they have no choice but to believe. At Enron many employees went as far as to put 100% of their savings into buying the stock.
Feels ironic now that Warren is himself an oil prospector with his Occidental stake.
This incest investing is pretty telling. The whole economy is about to get burned- doubly so with DJTs tariffs and bonds not looking like the safe haven they usually are during turbulent times.
“We’re now locked into a particular version of the market and the future where all roads lead to big tech,” says Amba Kak, co-executive director of the AI Now Institute, which studies AI development and policy. Indeed, the success of major stock indexes—and perhaps your 401(k)—is resting on the continued growth of AI: Meta, Amazon, and the chipmakers Nvidia and Broadcom have accounted for 60 percent of the S&P 500’s returns this year."
This is fueled by the parabolic token usage. If token usage continues to grow parabolic, then these are the most genius deals made.
If token usage declines because of high prices, look at what happened to SBF.
I notice that when I ask ChatGPT a question the answers I get back seem more verbose than they were a year ago. Where a 1-2 line response would be sufficient GPT delivers a sprawling essay. Overall the value of the answer has probably gone down, but tokens are up.
You can quite easily ask it to summarize the result in a sentence or paragraph. LLMs have no other way to compute than write text and the more text they write the more compute they do. You only care about the final output.
I do. It typically goes on to write a preamble about why it gave a long answer before finally providing a summary. Token stuffing.
This is akin to complaining about the overhead of radix conversion between decimal and binary in early computers.
Unless you figure out how to talk in latent space directly to an llm the preamble is the cost you pay for getting the answer back in English.
I've just set my (free) ChatGPT to "Robot" in style options, so far I like its style and level of detailedness
> look at what happened to SBF
SBF lacked funds to ride it out. If they survived for a bit longer, Bitcoin would have surged and they'd keep going. AMD, Nvidia etc all have income and funds to survive.
This is just about finances unless you're implying there's some crime here too.
> SBF lacked funds to ride it out. If they survived for a bit longer, Bitcoin would have surged and they'd keep going.
Yes and then maybe he'd never have been arrested and jailed for all his crimes.
For an interesting interpretation of the recent AMD-OpenAI deal, see Matt Levine's column from a few days ago:
> OpenAI: We would like six gigawatts worth of your chips to do inference.
> AMD: Terrific. That will be $78 billion. How would you like to pay?
> OpenAI: Well, we were thinking that we would announce the deal, and that would add $78 billion to the value of your company, which should cover it.
> AMD: …
> OpenAI: …
> AMD: No I’m pretty sure you have to pay for the chips.
> OpenAI: Why?
> AMD: I dunno, just seems wrong not to.
> OpenAI: Okay. Why don’t we pay you cash for the value of the chips, and you give us back stock, and when we announce the deal the stock will go up and we’ll get our $78 billion back.
> AMD: Yeah I guess that works though I feel like we should get some of the value?
> OpenAI: Okay you can have half. You give us stock worth like $35 billion and you keep the rest.
https://www.bloomberg.com/opinion/newsletters/2025-10-06/ope...
https://archive.is/tS5sy
Discussed yesterday:
OpenAI is good at deals - https://news.ycombinator.com/item?id=45493815 (40 comments)
Levine is fantastic on all sorts of bizarre financial dealings. Money Stuff is such a good daily read even though I don't work in a financial job
I would be convinced that he's some sort of super well calibrated financial article AI engine (due to consistency of output and breadth of coverage), except that I've been reading his column long before LLMs were on the scene.
There's no way someone would be stupid enough to part with $x worth of product and $ 0.5 x worth of stock in exchange for $x, is there? Honestly, it's stupid to even make such an offer. What prevents AMD from turning it around on them?
> AMD: Oh, since you put it like that, why don't you pay us $x for the chips and give us $ 0.5 x worth of stock? When your market cap rises by $ 0.5 x you'll get the money for those stocks. And hell, it might even rise further, giving you back some of the money for those chips. You're really robbing me blind with this deal, you'd be a fool not to take it!
>What prevents AMD from turning it around on them?
Matt's telling is a little facetious, so the exact negotiation probably didn't go like that. More importantly, OpenAI has better options. If they want to pay AMD cash for chips, they could do that without also giving up their equity. The same can't be said for AMD. They're the second choice when it comes to GPUs, and they're desperate for market share, so they're willing to cut deals like this. Without this deal OpenAI probably would have bought Nvidia chips with cash.
> NVIDIA intends to invest up to $100 billion in OpenAI progressively as each gigawatt is deployed.
AMD might do it just to try and be (stay) relevant in the market. Prior to this, discussions about AMD were slim to none, "just another inference card manufacturer". Now they're shining in the spotlight. Mission accomplished.
If you really think about it though, inference is the end goal for most AI models. So I don't see "inference only" being necessarily bad, especially if you get more bang for your buck.
Because OpenAI has the hype and delivers the pop, not AMD. An AMD investment in OpenAI would probably have had a similar effect on AMD, not an equivalent one on OpenAI. (Either way it's Ooh AMD is getting seriously into AI, awesome, suddenly it seems worth 100x whatever it was before!)
Replace AMD & OpenAI with some crypto company and Trump, hypothetically the latter doesn't really obviously get any significant benefit from the former's endorsement.
Oh this just seems like the typical financial dealings that City of London invented to obfuscating ownership of holdings half millennium ago. Some would event argue this obfuscation what transferred power from the military to bankers. And it is how we end up in this transnational corporation where the public really have no idea who is owning what and who is actually running the show here.
Now we are just transferring the power from bankers -> tech/AI.
This reminds me of the AOL Time Warner merger. "Back in the day", America Online (AOL) had a valuation large enough to pull it off:
After less than a decade, once ISP's became a thing, the AOL part of the new company shriveled into nothing and was spun off. It still exists, in name if not anything recognizable to what it once was.0 - https://en.wikipedia.org/wiki/AOL#2000%E2%80%932008:_As_AOL_...
Does he explain why the valuation of the company would go up by $78 billion on the public news that AMD agreed to give $78 billion of goods to OpenAI for free?
It validates that MI450 works. Jensen was saying other GPUs are so bad that companies wouldn't even take them for free and then OpenAI said no, we actually will take AMD GPUs for free. They're not that bad.
I’m not sure that agreeing to receive a product for free indicates much about the effectiveness of the product.
It's very bad
It's not really for free. OpenAI buys goods for $78 bn and maybe gets that back in stock options or maybe not depending on how the stock does.
As to the stock going up - we have a bit of a stock bubble thing going on.
It’s just AMD levering up on its on stock. Similar to MSTR. AMD is taking a loan against its own shares to build gpu’s hoping to rent them out via OpenAI in the future. No point in betting against it, because it can’t get margin called until OpenAI goes bankrupt. Which isn’t happening until OpenAI is worth $10 trillion first.
From the OpenAI side, it’s an amazing deal. $35B profit in one day… at a pe of 30, it’s already worth a trillion if it can repeat similar deals every year.
MSTR can't print more of what they are levering against.
People are bringing up the AMD deal but isn't that giving it too much credence? The deal hasn't had any material consequences yet apart from stock market fluctuations. The bigger problem for me is that AMD doesn't seem like is going to be a player of note in the AI sphere. So this deal like many other big money AI deals look like optics to me.
> AMD doesn't seem like is going to be a player of note in the AI sphere.
I thought AMD is well positioned in the inference space? They have high VRAM, somewhat high connectivity, already shipping pods that are pretty ok for inference? Training is still dominated by nvda and their CUDA moat, but there's an increased need for scalable inference, and that should fit AMD well. Am I wrong in thinking that?
That's what I have read as well but the ocean's distance between Nvidia and AMD in terms of market share of AI chips does not reflect that.
The MI450 is the pets.com from the 90s of AI hardware. It's bad.
The bubble now isn't even about LLMs, but has affected the entire stack. Even hardware companies are offering rubbish for the sake of prop'ing up their own valuation.
CUDA had decades of improvements, we may have AGI (tm) earlier than what it will take AMD to fix their broken software.
MI450 doesn't work well at all, OpenAI would rather take the deal and overspend in a project to try to use it for their software stack, which is incompatible with it.
Just for the sake of proping up valuations, with no "market competitiveness" guidance. Just for moving money.
Funnily enough, Google is ahead of all of this, as they can make their on TPUs and have their integrated stack. They don't need to generate fake economic activityu.
What is the economy if not a bunch of much larger overlapping circles of different sizes?
And I guess the smaller the cluster, the larger the bubble?
On the surface, the circular financing seems worrying but once you dig into it, it seems quite benign, tbh. NVIDIA is handing out GPUs into OpenAI, in return for stock/future revenue, whichever way you want to slice it. OpenAI has wonky unit economics, yes, but they're growing at ~100% YoY, so it all checks out, if you ask me.
> wonky unit economics, yes, but they're growing at ~100% YoY
Isn't this "losing money on each sale but making it up in volume"? Sounds like I heard that before and it did not turn out well.
That was Amazon's motto in mid to late 90s :). They almost went bankrupt. Now they are ruling retail and cloud.
On the contrary, Amazon could have become profitable on demand. They were just investing heavily into physical infrastructure back-to-back and if they ever wanted to slow down or stop, they'd be default alive.
Is more complicated than that.
Walmart back in the 80s investd heavily into physical infrastructure back to back - building new stores, their own trucks fleet. The first retailer with its own satellite communication system. There was never a question about Walmart going bankrupt.
I spent a couple of years in Amazon retail finance. The vendors were incentivized to sell as much as possible, under cost if that helps. Part of my job as financial analyst was to "convince" them it is a bad idea to sell under cost. Very hard and stressful work.
The part that I think may have some credence is that it may be a way to prop up huge profit margins that may persist for longer than they would otherwise in lieu of further commoditization and competition.
If NVIDIA has a stake in the company, they are less likely to do something like start brewing inference chips in house with the help of the foundry partner and a provider of vanilla chip designs. The company also gets a huge cash injection that is somewhat contingent on not doing that, and hey, they have a fresh pile of cash for cutting edge chips so whatever, right? My first thought was that these deals had more to do with that than anything else.
That aspect may end up being a bit illusory in the end. But then again, Nvidia has been proven to be quite skilled at building out and defending their ecosystem, sometimes through ruthless means, so maybe it persists (and I'm not sure that's a good thing). China certainly isn't a threat to throw a wrench in this situation... the entire US geopolitical complex will ensure that is the case.
The weird part is the _impression_ that $100B is being paid to a hardware company for chips, when in fact zero dollars are being paid to a hardware company and instead they are relying on public markets to spike the stock price based on potential future orders as if the demand is real.
The only reason this works is if we are the useful idiots buying up the stock.
yes but their growth isn't driven by actual profits. It's a bet.
A 0.1% tax on all money transfers (or any other transfers of value from entity A to entity B) would end this...
There are sales taxes on each sale (GPU sales / Cloud VMs sales / etc...)
Meta point: governments want to tax each money transfer - when people are paid, when selling a real estate property, when selling stocks, when selling goods and services. Every time money changes hands governments want to take a piece. This is the easiest way for me to understand taxation in pretty much any jurisdiction.
Why does "Web of circular deals" sounds like a scam?
Because it is.
"AI companies say these kinds of large, reciprocal, and sometimes overlapping deals are what's needed to meet growing demand for their tech. But as more and more money gets invested in the space these unorthodox business arrangements are also raising flags for industry analysts who are starting to wonder: Is the trillion-dollar AI market being propped up by the industry itself?"
From the outside: sure looks like. Can’t wait to find out if I’m right or wrong.
I think certain people are far too deep into it that there’s no rationale behind what is going on now.
It’s all about ‘accelerating’ forward at all costs.
Is this the proverbial writing on the wall then?
Every day a new deal involving OpenAI.
Every day one or more article wondering if/when the AI boom will transform into a bursting bubble.
Maybe people are crying wolf or misreading the situation.
My gut feeling is that yes, this is indeed a bubble and behind closed doors the CEOs are in panic mode, weirdly repeating past and well known mistakes
I think you could classify the recent deals as bubble inflating. We are probably a way off from it bursting.
It just we’ve seen the same thing for over four decades. I was getting stock certificates mailed to me from IBM in the 1980s when the savings and loan scandal rocked the market in 87. Then Enron. Then the internet bubble. Then the subprime meltdown. It just happens every decade or so when “too good to be true” runups happen.
Feels like some GPD hack where you could make it fake grow by just handing money around in a loop between actors.
Yeah I spoke to my wife a few min ago about these deals and other indicators of a bubble. We’re updating our 401k’s and the old college fund brokerage account in the morning and have agreed to not make any additional changes until Jan 2027. Going to sit out a year and see what happens.
Equities in the US are priced in dollars. If you think the dollar is going to lose value(because the government is led by people who have repeatedly told you this is their goal) then you expect stocks to rise commensurate with the devaluation of the dollar. I am not a bitcoiner or saying the USD is done for, but it seems clear that the dollar is weakened and will continue to weaken and this is a significant risk for folks like me with minimal exposure to stocks and precious metals.
Why would stocks raise with commensurate devaluation of the dollar? Stocks financial positions (assets / liabilities / cash flow) are most likely denominated in dollars.....
If companies' financial position would be denominated in something else (gold / Yen / Euros / CNY) this comment would make more sense to me....
Companies values are what they are. Equities are essentially their own currency. The denomination of their conversion price doesn't change their value. The listed price on the nyse or NASDAQ can be thought of the exchange price between an equity and USD.
If the USD loses half its value, the value of Microsoft does not change.
you understand that costs rise on devaluation of the dollar right?
as the usd devalues these stocks will also skyrocket profit/revenue
These sorts of outlooks are notoriously hard to predict.
Paul Tudor Jones (one of the most successful investors of our time) is of the mindset that if anything we are about to see equity appreciation far in excess of what we have seen for decades.
https://fortune.com/2025/10/07/paul-tudor-jones-hedge-fund-b...
If that's the case, right when you're looking to rebalance your portfolio, you're going to be looking at a much more difficult decision, having already missed out on price appreciation that would be quite useful in padding downside risk...
Perhaps just step away from market cap weighted index funds as a long term adjustment. That's something that with a proper basic framework is advisable as a general portfolio management approach as well. (I have some SMLF - Smallcap Quality/Value, Berkshire, generic Global Equity exposure, not for outright returns, but to lower exposure to the concentrated passive market cap weighted indexes and diversify risk exposures).
Trend Following exposure is a great add as well, since it is negatively correlated with risk assets in many macro/market regimes.
Probably healthier to address your very real concern with modifying the long-term portfolio design rather than take a short-term market timing approach which is more of a negative expected value.
> These sorts of outlooks are notoriously hard to predict
The gut feeling you're about to lose your life savings isn't worth it for a lot of people, when I had most of my money invested it was always in the back of my mind. I cashed out and bought land/started building a multi generational house. All doubts and fear disappeared overnight, I'm for sure losing money compared to going all in on Nvidia or even btc, but I don't need any mental gymnastics to rationalize my choice.
> I cashed out and bought land/started building a multi generational house
interestingly, my wife and I have discussed this a lot too. We have 20 acres in SE Oklahoma that we got a really good deal on during the pandemic, the land is unplatted and the realtor didn't know there was water and power service already setup by the previous owner. We want to build a small cabin/house over the next couple years. It's like a retirement backup plan, "if everything goes to hell we can still go sit on our land in OK and watch the sunset..."
> if everything goes to hell we can still go sit on our land in OK and watch the sunset..."
No you can't.
The sweet nuclear summer glow from the back porch is an ambiance simply unequaled.
There has been no point in history where anyone lost their life savings investing in equities.
Never touch your portfolio except to rebalance and only rebalance based on risk, not the market.
When the market goes down, delete your banking apps and reset your passwords to random characters. Never sell
Quite conversely, if your equity portfolio is going up tens of thousands of percent yearly, that's when you should be sounding the alarm...
Unless you plan on withdrawing in the next few years, you should actually just do nothing
my 401k is setup in pretty high risk ( according to Vanguard ) index funds. Now that i'm older and looking to retire in the next 10-12 years, the investments need to be re-balanced anyway. As for the brokerage account, it's just a vanguard sp500 index fund. I'm not super exposed to the mag7 but they're such a huge component of the overall sp500 that when the bubble pops it's going to affect the sp500 quite a bit. I was actually looking at some of the smallcap Vanguard funds available to me because they seemed to fair ok in the .com bubble burst while the sp500 and especially the nasdaq were hit hard.
You are being mindful of some real risks and concerns.
That said, there's no reason to have to feel like you need to take a stab at it totally by yourself. Heck, that's what I'm doing in the day to day - putting together a few of these portfolios for people. There are certainly advisors and managers who know the academic side of this and can diversify away from concentration risks and such.
There are also many great tools in the relatively recent future that have opened up for common investors to diversify smartly. You can get pure trend following wrapped up in an ETF. You can get 100% equity 100% bond wrapped up in an etf at half of the size, so you can fit Gold and Trend in a portfolio with no hassle (both are low return but great from a diversification standpoint, so most retail investors want to lever up a bit to get back to the standard SP500 return/variance profile).
If you don't or can't find someone who can math it out for your individual case, it's also an option to just pull a few of the best proven individual ideas, such as "have some international diversification". ETFs are easy to grab here. Even 10-20% can pad out some of the US concentration and dollar exposure a bit. Smallcaps, like you said, are another way to shift away from the "size factor" dominated index weighted ETFs. The Nifty 50 bubble in the 60s was a major catastrophe, and history rhymes, so why not hedge a bit while still sticking to a proven investment framework? No financial advisor is going to look at an equity slice of 80% S&P, 10% international-ex-US, and 10% smallcap and say it's a stupid idea. Actually, they will probably say it's a smart improvement over 100% S&P.
When it comes to smallcaps, the caveat is that the Russel2000 is filled with fairly low quality companies. High debt, value traps, you've got it all. That's diversification, but probably not the type you are seeking. That's why smallcaps are a great target to add some factor based investing returns. This is another diversified source of return in and of itself (AQR as 150 billion under management here), but why not overlay some Quality and Value factors on top of the smallcaps? Now you have a sprinkle of factor investing edge in the portfolio alongside the smallcaps, while not drifting too far from the status quo since smallcaps are a small allocation. Not going to toss out specific investment ideas but there are ETFs/mutual funds for this! Just stick to low fees and a long track record. A question like "I want 10% in smallcaps, but I want a factor overlay for Quality and Value", is a question that would actually get money out of a good advisor by the way.
Here's a good, slightly wonky, lecture on some of the esoteric yet quite "in your face" risks within a complex dominated by passive and index weighted flows. It's not at all a misguided though to think there are some risks out there in the left tails. This was a lecture from 5 years ago: https://www.youtube.com/watch?v=x-rJciYZmi0&t=1s
If we end up with inflation rather than a market crash, this could be quite costly? Gold is up 50% this year, which hype is more correct than the other?
Gold is up because the dollar is down.
Like guys... Gold is always worth the same. It's the other currencies that change in exchange rate with gold.
I am for a medium risk profile and I've already started diversifying. A lot of major "global" or "index" tracking funds are now majority comprised of holdings in Nvidia, Microsoft, Meta, Alphabet and also Tesla. I've seen many supposedly non-tech funds where about 50-75% of their portfolio are linked to this bubble. Sometimes the holdings aren't clear if the fund invests in other funds. As an individual investor I won't be able to react fast enough to a crash, so am being proactive with moving investments so that no more than half is exposed. Will obviously keep an eye on things to reduce that if it looks wise.
There is some circular financing going on, but AI accelerationists think this will be offset by demand, value, and adoption in businesses. Hence these deals are warranted for the incoming demand.
I think I saw Altman saying there's a global shortage of compute just now so this may address it. I'm not sure how much is actual user demand though and how much just investors wanting to pile into AI startups.
Upvote because can someone explain to someone as dense as me, whether or not this is likely to make some likely AI bubble worse? Is this just how industry allocates capital?
Part of what's concerning here is that the deals are conditional. OpenAI must meet XYZ conditions before cash/stock/etc is transferred, and the conditions are pretty hard to meet.
The money between OpenAI, Nvidia, Oracle, AMD is not circulating. There is no cashflow, only future commitments that may (and quite likely will) collapse. Yet the stock market & media react as if it's a sure thing. Even in the criticisms of these deals, the hype is affirmed.
This is the same problem as Enron's accounting, minus the fraud. (No need for fraudulent accounting when people simply don't read the fine print.)
These deals certainly make the bubble look larger because people are double-counting revenue. They also seem to be triggering extreme investor FOMO.
> They also seem to be triggering extreme investor FOMO.
Bubble's only bad if you get out at the wrong time.
Only for you as an individual, from an economic and societal perspective a bursting bubble is never good!
It’s not the bursting that is bad. It’s the bubble that is bad. The bursting is unpleasant, but good.
Much like a boil.
The money is allocated by institutional investors. They buy the stock, the companies trade their more valuable stocks amongst themselves for various deals. If institutional investors stop investing then the flow of deals stops & the bubble pops. There is nowhere else the institutional investors could park their money other than tech so I don't think the bubble is going to pop any time soon. But infinite growth is obviously a logical & physical impossibility so eventually there will be a correction but whoever says they know exactly when that will happen is lying & they'd be better off buying lottery tickets to cash in on their ability to predict the future.
Personally think this makes a tech bubble a contagion for other parts of the financial industry - especially if institutional investors take the "easy" trade that everyone is doing and add leverage to it.
Now once these folks don't get 800B per year in revenue and the money runs out, all of the banks go as well. But don't worry - they'll get bailed out with our money...
>AI bubble worse?
Is it worse than the dot-com bubble? I remember everyone and their mom who knew HTML could get hired, and there were way more companies that went IPO during the dot-com bubble, like theglobe.com.
AI is a bubble, but is it worse than the dot-com bubble or the real estate bubble in 2008?
> Is it worse than the dot-com bubble? I remember everyone and their mom who knew HTML could get hired, and there were way more companies that went IPO during the dot-com bubble, like theglobe.com.
IDK, the number of companies and employees might be smaller, but the valuations and comp packages are so much bigger that I suspect there's more money in the bubble this time around, just a bit less spread out.
The dotcom bubble had investor capital going towards hiring tens of thousands of people. The AI bubble has capital going towards buying hardware, signing a handful of researchers on 9-figure deals and settling lawsuits for IP theft.
> remains largely unproven as an avenue for profit-making
That seems like a bit of a stretch. How much money is OpenAI making?
Pretty sure they're losing money on the products that they actually sell. All the cash they have is coming from investors (ie hype).
If memory serves, it's shit like this that sank Global Crossing: creative accounting, I think it was called at the time.
If the AI bubble bursts, there's a high chance the Everything Bubble bursts right after. And then USD goes gg.
I think part of the problem is investors are counting on infinite bailouts because AI is too big to fail. But that also causes USD erosion.
I have no clue what is the way out no matter who is in the driver's seat. And speaking of that, we had a decade of incompetent people in key positions at the Fed/Treasury, BoE, and CBE.
But "nothing ever happens".
Circular trading is a common technique used in Ponzi schemes to create the illusion of trading activity and to attract more investors.
This and stock buybacks make me question company valuations
Stock buyback is effectively just a dividend with a different tax implication: reducing the number of shares in circulation raises the ownership stake of the remaining shares
> Stock buyback is effectively just a dividend with a different tax implication
Not just different, it specifically a lower tax rate assuming that the stockholder has held long enough to use the long term capital gains tax rate (which lower than the dividend tax rate).
[in the US]
Stock buyback isn't a total scam as it seems, but it does mean "we can't figure out any productive use case for this cash in advancing R&D or scaling our business anymore" which is still pretty worrying
By that logic, any profit a company distributes to shareholders is worrying.
I suppose stock buybacks are similar to dividends in that regard
It’s more like, “the executives with lots of shares can’t see how to make the company grow, so they’ll just use profits to pump up the share price for their gain”.
I deeply feel buybacks shouldn’t be illegal but treated shamefully.
Instead of using profits to build up long term savings or fund R&D, they basically choose to do as little as possible.
There is no vision.
Realistically, if the execs think they can't do R&D then they can't. If they tried they would just waste the money.
> Instead of using profits to build up long term savings or fund R&D, they basically choose to do as little as possible.
Isn't buying stock a form of long term savings? After all, they can always sell that stock when they want to "withdraw".
Sure, they may not get the same return as simply stashing it into a bank account, but it's also a statement of confidence in their future: "We are sure that our stock will outperform every other option there is for storing our money, because our long-term plans include extracting more profit from the market."
Should companies never pay dividends?
Investments into vision can juice the stock price too. So they would do it if they had good realistic inspiring ideas
clearly openAI is the top dog here who can hoover all the money , but it's not a public company
So far OpenAI hoovered - (minus) $7 Billion in first quarter 2025. $11.2 billion OPEX (compute, sales/marketing) in exchange for $4.3bn revenue. I wonder if revenue sharing agreement with Microsoft means Microsoft gets to pay for half of that :-)
I do wonder a bit about those finances. Losing $7bn on 4bn revenue so let's invest another $1000bn capex in this wonderful business.
Here is money printing scheme that looks to be at work:
Initial situation:
* Big corp M has X$ in cash where X is huge
Big Corp M invest X$ in AI startup O, with a provision that O needs to use most of the money to buy cloud infrastructure from M to power AI models.
End situation:
* Big corp M has X$ wort of shares in O, the value of which will rapidly grow
* Big corp M cloud division has ~X$ in extra revenue
The deal automatically turns X$ into ~2X$ in books. Rinse and repeat with next round deals and next AI startups. The big corps are reporting increased cloud divisions revenue from AI spent, but it is their own investment money flowing back to cloud divisions.
> The deal automatically turns X$ into ~2X$ in books.
No it doesn't. The revenue it gets back is valued only at a fraction, because it's only worth its profit. Revenue != profit.
And your "the value of which will rapidly grow" is doing all the work here. That's not guaranteed. It might collapse as well.
It's not money printing at all. It's tying up cash long-term in exchange for a much smaller amount of profit short-term. Which is risky but entirely normal.
Nothing "money printing" about it.
To be precise I wrote 'in books' which record also revenue, not just profits. Increasing cloud revenue is one of the things that drives big corps share price up, the missing bit is that this revenue comes from big corps themself. The growth investors mantra is that as long as revenue is increasing rapidly, they don't care much about today's profits, this is why so many unprofitable companies have crazy high valuation.
All I want to know is, should we be investing into these companies as a result of these deals? Or should we be moving out of these positions.
The OpenAI/AMD deal has an explicit AMD price target of $600. I don't see how it can be any clearer.
Not possible for AMD to get there just from this deal alone.
All I want to know is, is the bubble going to pop next week, next month, or two years from now?
Welcome to the casino! Please enjoy your stay, and remember: the house always wins ;)
It’s not a casino, I’ve been winning for years.
By asking random people on the internet what to do?
Yes, many recommended good index funds or FAANG companies.
Keep in mind that it’s easy to win in a market that’s just going up. You could pick random stocks and you’d probably get decent returns.
The last decade has been pretty spectacular for stocks but that doesn’t most of us wizards at investing.
The thing is if you’re well diversified, then when you win, you win big, but when you lose, you lose small.
My worst losses have left me with maybe 0.5x-0.25x of my inital investment. But my big wins have been 10x+.
Even if you leveraged hard, brokers would close out your positions before you go negative.
Never heard of a winning streak at casinos?
that doesnt make you nearly as special as you think
We are fully in the grifting stage of society. No one wants to work anymore and everyone wants to invest in imaginary outcomes (cryptocoins (most of which no one uses with little to no real use cases) and now AGI (science fiction)). Once the stock market crashes within an already low trust brittle society I imagine that there will be catastrophic outcomes. Maybe I am just a doomer who needs to touch grass but maybe not.
“Web of circular deals”
Welcome to how economies work. You only worry when you can’t see where the ends of the circle meet up. Ponzi schemes are notable specifically due primarily to the fact that they’re so directional and one-way. If you can’t pinpoint the escapement and see how it’s feeding into the winding of some other spring, walk away.
What economy has this much money moving between 7 companies across such a narrow set of operations?
“Web of circular deals”
Welcome to how economies work. You only worry when you can see where the ends of the circle meet up. Ponzi schemes are notable specifically due primarily to the fact that they’re so directional and one-way.
Cyclical growth
Really? https://news.ycombinator.com/item?id=45511368
My brain now reads Sam's face as an "AI generated" watermark.
> Two weeks ago, Nvidia Corp. agreed to invest as much as $100 billion in OpenAI to help the leading AI startup fund a data-center buildout so massive it could power a major city. OpenAI in turn committed to filling those sites with millions of Nvidia chips. The arrangement was promptly criticized for its “circular” nature.
That's not a "circular" deal, that's a... regular deal.
I genuinely don't understand the criticism here. Every business deal is -- I'll do something for you, you do something for me, and we'll both be better off. That's the free market.
Regardless of whether AI is a bubble, this is just business as usual.
A healthy two‑way exchange isn’t the issue, as you note, partners routinely buy from each other. The problem critics flagged here is that this particular arrangement starts to resemble “vendor financing” or a circular funding loop. Nvidia commits up to $100 billion to OpenAI. OpenAI then uses that money to buy Nvidia’s chips. Nvidia books the revenue and likely earns equity or upside in OpenAI. On paper, both sides look like they’re growing dramatically. But it’s the same capital being passed around, and it makes it harder to tell how much real, third‑party demand is driving those numbers.
Why people worry:
• Real demand vs. vendor financing: When the supplier is effectively fronting the cash for the customer’s purchases, it’s not the same as independent spending from end users. Investors and regulators want to see whether OpenAI could buy all those chips without Nvidia’s financing.
• Confusing financials: Nvidia appears to rack up huge sales, while OpenAI’s balance sheet swells with cutting-edge hardware financed by Nvidia. Untangling how much of each company’s “growth” is externally driven versus funded by the other side becomes tricky.
• Risk concentration: Nvidia’s fortunes would hinge on OpenAI delivering an multi-trillion-dollar scale business to justify the investment. If AI demand doesn’t materialize as hoped, Nvidia isn’t just missing chip sales, it’s directly exposed to OpenAI’s downside.
• Governance & optics: When vendor and customer roles blur, people question whether decisions are being made to satisfy true market demand or to juice financial metrics and valuations.
So it’s not that reciprocal deals are bad; it’s that circular, high-dollar deals can obscure what’s actually happening under the hood. That’s why folks are raising eyebrows.
It’s like Sam giving Pat a $20 “investment,” and Pat immediately spending that same $20 to buy Sam’s lunch vouchers. Sam says, “I sold $20 of lunches!” Pat says, “I’ve got $20 in lunches!” But really, it’s one $20 bill just doing a lap.
I'm generally of the same take as you: this reads much more to me like 2009-2013 Apple where they fronted cash to suppliers to do build outs and get to the 10x scale Apple needed ASAP. People looking at this as "proof" there's a "bubble" scare me.
HN community predictions so far are
2000s: cloud bubble 2010s: crypto bubble 2020s: AI bubble
Seems modern economy works different. I missed cloud and crypto and dont wana miss AI so im investing about 20% of my income.
thats sort of what you get when you take a bunch of developers opinions as facts on an ever changing global economy.
a lot of comments dont really understand finance; and are just repeating gut feelings they align with
How are these "circular deals" different than Boeing (and other heavy civilian equipment makers) providing financing on purchases? My point: Is this simply clutching pearls over the AI/LLM bubble, or is there some substance to this concern?
And another: How does it compare against the new breed of Bitcoin treasuries, like Strategy (MSTR.OQ)? As I understand, Strategy uses secondary offerings to continuously see more shares, then uses the capital to buy more Bitcoin.
This happens also with automotive companies. Peugeot has its own bank, which lends money to people buying peugeots. The big difference: is not circular. The bank is a bank, the OEM building cars build cars.
The bank makes money by lending and cashing interest.
The car maker by selling for profit.
The money goes from bank to OEM, and not the way around, no circle.
I do not know the specifics of Boeing though.
Boeing uses Export / Import bank to finance some of its sales:
https://www.exim.gov/news/export-import-bank-united-states-b...
https://www.reuters.com/business/aerospace-defense/us-export...
https://www.cato.org/regulation/fall-2015/us-export-import-b...
Didn't Boeing's deals turn out to be bad?
how long until this thing pops?
my money is on 18 months
18 days
I sense a business opportunity here: betting on stocks! Maybe we can have a central market where people can exchange their money for bets. Maybe even sell their bet off before the due date.
We can call it a "stock market", or maybe even a "stock exchange". I expect I'll be able to get a patent on it if I prefix the name with "online", or suffix it with "on the internet".
Even better, form pairs of stocks and make them fight! Bet on the outcome using any liquid electronic currency, such as CS skins!
Anyone remember the subprime crisis?
this is going to be bigger, mostly because the leadership vaccume in the US Gov is all too happy for it to get very big before it goes bang.
The other day I was helping somebody choose some funds to invest in. The majority of "global" or "index" funds we looked into either directly or indirectly had the largest share of holdings in Nvidia, Microsoft, Meta, Alphabet, AMD. Often these were about 50-70% of their portfolio. The returns for the past year look great but the crash is going to be devastating and widespread because so many investments link back to this bubble.
I completely withdrew from the US market last Nov 5th.
It's plainly obvious where all of this is going.
Where did you put your funds? bonds are screwed but I'm thinking gold
There are plenty of non-US index funds, like EU, UK, Japan and others. There are also indices that track smaller companies rather than just the S&P500 or Nasdaq.
Or diversify in both directions - small US, and big and small international funds.
There are other companies than the magnificent 7. Other boring companies with boring dividends and solid earnings that have nothing to do with tech. Ulta. Pepsi. Chevron. But you sound pretty smart so I’m sure you had a good reason to invest in non us markets.
I'm a non-american and I don't trust their government as a steward any longer.
I've diversified into international markets that are more resistant to the whims of a dictator and his feeble attempts at monetary policy.
Also any gains in dollar denominated assets should include the >10% haircut the USD has taken since the fascist coup.
well, this means you've missed out on everything since then, wouldn't a simple stop order have been a reasonable alternative on this?
the US dollar is down over 10% since they elected a fascist, no stop order for that.
[dead]
[dead]
[dead]
[dead]
Just enjoy it if you can’t avoid it
How about giving Nouveau some actual support so you're not screwing over everybody using Linux
They are doing that now. https://www.phoronix.com/news/NVK-Vulkan-Red-Hat-NDA-Docs
I can't legally define it yet, but I struggle to see how these deals aren't tax evasion, if not outright fraud.
That's not how taxes work. This isn't reddit, not everything is tax evasion.
OpenAI is spending little to no money on the hardware and AMD can claim both investment expenditure and revenue. I made clear that I'm not sure the intricacies of how that or may not be tax evasion or fraud, though I dotn see how it could be neither.
Would profiting from these deals make your struggle a little less difficult?
I don't understand why they are calling these deals circular. OpenAI is buying Nvidia and AMD chips. Oracle is also buying Nvidia chips. OpenAI is buying datacenters from Oracle, which will be powered by the chips Oracle buys from Nvidia. This is one directional: hardware makers (Nvidia and AMD) sell either to datacenter makers (Oracle), or to AI firms (like OpenAI). That's it. No circular deals.
But "circular deals" has such a nice ring to it, that you hear it everywhere nowadays. People are just hungry for negative soundbites.
I prefer https://www.bloomberg.com/news/features/2025-10-07/openai-s-... which shows the circularity in the deals.
A stronger counterpoint to the circular deals suggestion is at the end of the article, which reads
> Michael Intrator, CoreWeave’s CEO, acknowledged the circular financing worries in a recent interview with Bloomberg News, but said the public concerns will dissipate as more businesses adopt AI.
> “When Microsoft comes to us to buy infrastructure to deliver to its clients who are consuming 365 or Copilot, I don't care what the narrative is about circular financing,” Intrator said. “They have end users that are consuming it.”
It's on you to decide then if end users are actually getting value / paying for AI products.
> It's on you to decide then if end users are actually getting value / paying for AI products.
they're not
https://www.perspectives.plus/p/microsoft-365-copilot-commer...
"Brutally honest and cheerfully sarcastic insights on Microsoft’s low-code and AI technology, business apps, and the business of software - written by Jukka Niiranen"
What credibility does this guy have? Sounds like someone that has an axe to grind.
He has been working in IT and business for a quarter of a century, and the last fifteen years as a MICROS~1 consultant.
https://niiranenadvisory.com/about
Sounds like the circle was broken with a Ponzi scheme in the middle…
“When Microsoft comes to use….” is he counting chickens before they hatch?
https://openai.com/es-419/index/openai-nvidia-systems-partne...
OpenAI => nVidia => OpenAI. That's pretty circular.
In the end with most of these deals its the shareholders paying through market cap dilution. Given the current market structure (the big companies are tech companies) there's PLENTY of capital for OpenAI to fund their expansion.
They've discovered a cheat code IMO. Instead of using and raising money themselves, use their reputation/popularity and use their suppliers market caps (e.g. NVIDIA, AMD, etc). The deal makes sense as long as the value projected to be added (i.e. via efficiency gains, loss of jobs, changing society, etc) exceeds the capital dilution for the supplier; they use their equity but the leftover equity value increase makes up for it.
Given all the passive investing, and funds invested in the top tech companies this is a VERY large pool of capital. It however increases leverage if the value doesn't materialise.
The systemic risk is real
Nvidia is also investing hundreds of billions in OpenAI.
Yes. OpenAI is also investing in AMD. This was discussed yesterday on HN, following some very good explanation by Matt Levine at Bloomberg. This is a way for one party to reduce some risk, by enjoying some upside in the equity of the counterparty.
But this is not circular. Circular would be if I sell you an apple worth $0.25 for $2.00, and then you sell it back to me for $2.00, or other similar amount, and I get to mark all the apples in my inventory at $2.00 and show a huge profit (on paper). One can create variations of this blatant deal. Like I sell you some rubber for 10 times the market price, you make a balloon and then I buy the balloon for 10 times the market price. I may not have other balloons in my inventory, but plenty of rubber, and I show some nice profit. One can imagine other, fancier deals.
But in the case of AMD and NVidia, and OpenAI and Oracle, the direction is clear. OpenAI has a clear need for compute. They can buy it directly from NVidia and AMD, or indirectly from Oracle. They can buy it with hard cash (of which they don't have that much), or with their own equity, or some form of deal that offers the seller an upside in OpenAI's equity.
But there is not back and forth buying of the same item, or of rubber/balloons. All the deals seem legit. Is it possible that all the future compute will not be needed, because the AI craze will fizzle. It is, lots of things are possible. But that's general business risk.
> But this is not circular. Circular would be if I sell you an apple worth $0.25 for $2.00
When you draw a circle on a piece of paper do you put your pencil to the paper, start drawing a curve, stop and write the word FRAUD, and then complete the curve? Or do you just draw a circle?
I’m assuming that people are saying “circular” in that it looks like the money goes in a circle. (For one example) Nvidia invests in Lambda, Lambda buys GPUs from Nvidia, Nvidia leases GPUs back from Lambda, Lambda uses the revenue from leasing the GPUs to Nvidia to raise debt to buy more GPUs from Nvidia…
This is how financing works. When you buy a house, you get a mortgage from a bank. It is unusual to get a mortgage from a seller. It would feel a bit circular, right? But that is exactly what happens most of the time when people buy a car. They get a loan from the same company that sells them the car. Is that circular?
When you replace people with companies, the financing can become much more complex. The example you provided with Nvidia and Lambda seems quite reasonable to me. Here's an example that happens every day in the world of housing: banks lend money to house buyers. Then they package the mortgages and sell the resulting mortgage back securities. Then they take the money from the proceeds, and give more loans, and package them and create more mortgage back securities. Seems circular, right? But that's just how business is done. There is no Ponzi aspect to this, or fraud, or smoke and mirrors. It's just every day business. Nobody labels that as being circular.
When people buy a car they sometimes get a loan from the auto manufacturer’s financial services arm. What they don’t usually get are warrants struck at a penny for 10% of the manufacturer.
You sound like you know what you’re talking about. I only think I know what I’m talking about. Help me understand: What am I missing in the OpenAI / AMD deal that makes it non circular?
OpenAI: We need lots of GPUs, you make GPUs, but your GPUs are not quite equal to Nvidia's GPUs. And we are a growth startup, we don't have a lot of cash, can you give us a hefty discount on those GPUs?
AMD: We need cash too, you know? We'd love to sell you those thousands and thousands of high end GPUs, that would cover some of our R&D, and we could one day match NVidia. But we don't swim in cash either. We can't give you the discount.
OpenAI: What can you give us? You must be able to throw in something there. Otherwise, honestly, we can't make the deal.
AMD: What if we give you some equity? And if our deal goes well, and our GPUs start being viable alternatives to Nvidia's, maybe we'll be able to get close to Nvidia's market cap and even surpass them, just like we did with Intel.
OpenAI: Brilliant. We love it.
AMD: Yes, but that equity will be contingent on how the deal goes.
OpenAI: Sure thing, we'll take that.
> You sound like you know what you’re talking about.
Honestly, I'm not an expert either, but I've run a company, and I can all but guarantee that credit_guy above really does not know what he is talking about.
what is incorrect with what they are saying?
> what is incorrect with what they are saying?
I've replied else-thread, in detail, on exactly why his analogy to mortgage and car loans are incorrect.
His main point that "these deals cannot be circular because mortgage and car finance are not circular" is incorrect. The mortgage and car finance deals are not analogous to the Nvidia/OpenAI or AMD/OpenAI deals.
Everyday business isn’t based on hype.
The AI startup valuation largely is. I feel it quickly becomes circular because people make projections purely on other projections since the world is too impatient to wait and find out.
A single hamburger store is never going to be projected to have a billion dollars of revenue because people understand the total addressable market. Doesn’t matter how good the burger is.
The AI stuff is too new that people don’t have a firm grasp on the costs and profit opportunities. They don’t really even know how to define the TAM. Too many unknowns. Entire classes of labor could be replaced by AI —- or perhaps not.
With little grounding, it quickly becomes a circle of hype.
I don’t know what to tell you guy, but when people see money moving in a circle in a deal there is a good chance that “circular deal” might pop into their heads. Because it’s a deal that is shaped like a circle.
> Is that circular?
Doesn’t really seem so because at the end of the day the money goes from me to them. I don’t get my money back, I get a car in exchange for my money.
Also this deal didn’t begin with the manufacturer purchasing shares of me before offering me debt to buy a car from them.
>But that is exactly what happens most of the time when people buy a car.
Your car manufacturer leases your car back from you? And you use that revenue to raise debt to buy more cars from them? What manufacturer are you doing this with? What do you end up driving?
> Your car manufacturer leases your car back from you? And you use that revenue to raise debt to buy more cars from them?
Honestly, if this was possible people would be doing it (not that they are not - fleet services and rental fleet services do some pretty funky accounting sometimes, so I wouldn't be surprised at all if this sort of thing happened).
If this was possible, I'd be doing it.
> What manufacturer are you doing this with?
Good question. The minute an answer gets posted I'm going to have a really good side-hustle.
> What do you end up driving?
I presume, at the point that this is being done, there is no actual car involved, so nobody involved will be driving because there is no car to drive.
> Good question. The minute an answer gets posted I'm going to have a really good side-hustle.
Yeah “everybody buys cars through the infinite car glitch” sounds like the sort of thing that would be part of an enormously long answer to “should I try WoW for the first time in 2025”
It's ironic that you described how the 2008 financial crisis came to be to illustrate how “normal” this circularity is.
Loans have happened long before 2008 and have continued ever since.
This process is described in the Bible for example!
What does the Bible say about collecting interest on loans (the necessary part for making mortgage backed securities)
"Do not charge your brother interest on money, food, or any other type of loan. You may charge a foreigner interest, but not your brother." - Deuteronomy 23:19-20
I also like Ezekiel 18:13
I sort of would read bible passages describing modern financial instruments.
Perhaps it has some psalms discussing merits of MMT, or parables on Quantitative Easing.
> But that is exactly what happens most of the time when people buy a car. They get a loan from the same company that sells them the car.
No, they don't!
Serious "Citation Needed" here. They get a loan from a financial services company, that is a separate company from the automaker and/or dealership.
The certification and requirements for trading as a credit provider will not be met by neither the auto manufacturer nor the dealership.
> Here's an example that happens every day in the world of housing: banks lend money to house buyers. Then they package the mortgages and sell the resulting mortgage back securities. Then they take the money from the proceeds, and give more loans, and package them and create more mortgage back securities. Seems circular, right?
No, it doesn't! They make out new loans, sure, but they aren't loaning you specifically the proceeds from the sale of your specific mortgage-backed security!
If you happen to get a new loan from the sale of the MBS, it is impossible that only you get that loan, from the sale of an MBS that had only your existing loan.
Seriously, there's laws and regulations around this, and from what you say, with all due respect, it seems that you are unaware of these regulations.[1]
The only reason that these actually-circular deals can go on right now is because OpenAI (and other providers doing similar circular deals) are not publicly traded, and thus there are fewer regulations and even fewer enforcement of what little regulations there are for unlisted companies.
=================
[1] Why is your handle "credit_guy"? You don't appear to be familiar with the fact that credit providers are heavily regulated in all jurisdictions that we are talking about. I mean, you don't even need to know the specific regulations and certifications necessary, you just need to know things like a dealer cannot be a credit provider too.
> But there is not back and forth buying of the same item, or of rubber/balloons. All the deals seem legit.
Some, certainly, but the vendor-financing deals certainly look circular to the casual observer. Microsoft invests $X into OpenAI for a 51% (or whatever) stake, and that investment then goes straight back to Microsoft to pay for compute credits.
Or Nvidia invests[1] $100m into OpenAI, which OpenAI then turns around and pays back to Nvidia for compute.
The majority of the deals making the news are structured like this; maybe technically those aren't actual ducks[2], but they sure look, walk and talk like ducks.
Similarly structured deals are with Oracle. And Coreweave. And everyone.
It may not be a "circular" deal, but what do you expect people to call it when a company makes a deal to receive cash (not credit, but actual cash) from a vendor, and spends that cash with that vendor?
==========================
[1] I use this word loosely here - the investment is a commitment of 10x $10m tranches.
[2] I.e. circular deals.
I agree with your general point: there is a tendency to group not least these deals together without distinction as confirmation of a pre-existing belief that the AI bubble is soon to collapse. The latter certainly might be true, but we should be careful about too quickly deciding what evidence supports it. I am inclined to strictly separate: (a) the NVIDIA deal as a vendor subsidizing one of their largest clients, which perhaps signals market weakness; from (b) the AMD deal as a non-market-leading vendor trying to entice a new enterprise client, who will only be able to use AMD's chips for serious training after significant (and risky) collaboration to improve their product offering.
This distinction is important to me because I see more concrete evidence for a possible material drop in NVIDIA's business short- to medium-term than a collapse of the sector as a whole. The chips act clearly shut them out unexpectedly from their second largest market and now it seems likely that Chinese chips will be competitive for training sooner than expected. Indeed, if Chinese AI firms are suddenly able to obtain even a roughly approximate product at considerably less cost, OpenAI will suddenly find themselves at a cost if not compute disadvantage to their Eastern competitors. It isn't a surprise, then, that OpenAI is now looking to reduce their present vendor lock-in with NVIDIA.
Overall, it's not great for the latter if they lose access to their second largest market, suddenly have viable competitors in their home market, and either lose some of a major client's business or have to significantly reduce pricing to retain it.
[dead]