When there’s a bubble, at first you hear “hmm this is weird maybe it’s a bubble”. Then more people start saying ,”yeah it looks like a bubble”. Then more people start analyzing how it IS a bubble. All the while big investors are like, “ I know it’s a bubble but right now I’m making bank, so…”. Finally, after those investors decide it’s been a good run, they cash out and the bubble truly starts bursting.
So right now everyone knows it’s a bubble. What we’re seeing is the big investors trying to squeeze every last billion out of it.
Finally, after those investors decide it’s been a good run, they cash out and the bubble truly starts bursting.
This time they wasted so much money that they’re trying to foist the bad investments on retail investors with overblown valuations and IPOs before cashing out.
You seem to be talking about SpaceX, but I gotta point out that with the fast-track it’s not just the retail investors holding the bag. It’s anyone with a retirement fund.
it’s not just the retail investors holding the bag. It’s anyone with a retirement fund.
I was thinking retail investors — whether individually or through an index fund — covered both. If you have any control over your retirement funds though, you can use that control to get out of the stock market if you want. So it isn’t anyone with a retirement fund, just nearly anyone with a retirement fund.
He didn’t make it into the s&p 500. NASDAQ bent over and opened wide though.
Yeah, heard it all before, and I’m very familiar with the structural “curiosities” of the existing investment landscape.
Very few people correctly called the problems with 2007-2008. Not none, but few. And with soooooo many people mindlessly on the “it’s a bubble!” bandwagon so early, a lot of accuracy and legitimacy is lost months or years beforehand for no other reason than why conspiracy theory people say “we’ll get UFO disclosure this year!” Or “This year the Cubs/Arsenal/Red Sox will do it!” It’s just the thing they say until one time they’re right.
I’m not telling you it won’t happen in a sense… But it’s not going to happen how or when you think. IMO, you’re looking at a partial stuttering effect maaaaaybe late winter like Q1 2027, and that’s about it. There’s to much alternate demand for everything LLM companies are already buying up to create a full and similar bubble like the Dot Com bubble.
You’re only cherry-picking some panic-mongering bias-confirming reporting here. Honestly - Yahoo? NBC? CNN? Why not Grandma’s Facebook shares and the Babylon Bee?
Plus, did you even read the Forbes article? The opening paragraph is literally what I’m trying to tell you.
For example, one huge thing you’re missing this is that the large companies of the Big 4 with frontier models aren’t taking out loans for this. It’s all the startups that are trying to draft on them that are using loans. THAT is a danger, that tens of thousands of idiots will sink a corner of the financial system from the sheer weight of defaulting loans because ChatGPT said their shit sandwich idea was “revelatory, and honestly, a great idea!”
But that’s not going to tank OpenAI or Google. Google has cash money. OpenAI as a fair bit of cash money. Meta has cash money. Anthropic had cash and then Trump tried to sink them… but they might be OK in the long run. They’re doing great on the code side of things.
What you’re also erroneously assuming points to a bubble points to how OpenAI, Google, Whatever Musk calls his stupid company this week, and Palantir all are doing to avoid classic bubble economics of huge loans to pay back. You don’t pay back stocks. It’s equity. That’s cash trading hands, not loans that come due one day. Hell, Antropic is buying server time from Grok - that’s real cash trading hands to perform a service. That’s not a bubble, that’s the kind of thing that prevents a bubble. Like, bro, do you even bubble?
I don’t claim I know when the correction will happen but I wonder what massive alternative demand you see for the mountains of highly specialised server gpus in storage that will be obsolete in maybe 3 years time? For many of them that means likely before they will ever be turned on. The dotcom bubble created infrastructure that was, largely, not obsolete when the bubble bursted and made a lot of sense to salvage. That is a fundamental difference to inference infrastructure.
The WorldCom fiber layouts were akin to the railroad bubble in the 1840s, in that those were pathways with nothing to use them. I can see the parallels here, but the difference is GPUs aren’t nailed to the ground. They can be moved to demand, unlike railway lines and fiber lines.
GPUs process data. They don’t spoil or expire. Sure, they’ll lose value, but it’s not like they stop being useful, even if highly specialized. Hell, even selling them second hand to China with an export waiver would be a way to recoup value. So already, the premise is flawed in that, specialized or not, China will use them. Or the EU or universities looking for a deal and building out their own local processing.
Both rail and communication infrastructure lead to some useless connections but much of it was no useless, in both cases. GPUs are not bolted to the ground but they do become obsolete no matter if you deny it or not. The issue is that the real costs is in using GPUs is very different from these previous bubbles. Those obsolete GPUs will cause much higher operating costs than newer generations, to the point where they won’t be interesting to use even if you gave them away for free. To make matters worse, other infrastructure is much more flexible in its use, one can transport all sorts of things on railways, one can send all sorts of data on communication infrastructure. Those specialised GPUs aren’t very useful for anything other than a fairly narrow use case.
I think you do not fully appreciate the crazy amounts of GPUs we are talking about here. China has no massive real shortage of GPUs. They managed to get black market GPUs more or less directly from Nvidia just fine. Nor are European universities IT wastelands without compute capabilities. But even if they’d go crazy on expandig compute infrastructure with outdated power hungry GPUs, that would be barely more than a drop in the ocean. Nvidia does have to resort to circular financing to keep the boom cycle accelerating, with GPUs going just to some storage facility if they exist at all. That is not how healthy demand looks like.
If you are talking about GPUs being only a small share of the overall total sum, bad news that the supporting infrastructure is also to a large extend tailor made for that very narrow use case. No one else will need such huge data center facilities designed specifically for GPUs, that includes also the non GPU components. And the infrastructure is the only thing of substance of this bubble. The models aren’t it. Open weight models are on the heels of the closed models. As soon as they are good enough for common applications, the business case for charging billions is slowly evaporating.
You are also mistaken, I am not worried about GPUs. I am merely stating that they and their server infrastructure (which is tailor made for them) are rapidly getting obsolete equipment by their nature and while the clock is ticking they are largely not even being used. This is fundamentally different from the dotcom and railway bubble.
When there’s a bubble, at first you hear “hmm this is weird maybe it’s a bubble”. Then more people start saying ,”yeah it looks like a bubble”. Then more people start analyzing how it IS a bubble. All the while big investors are like, “ I know it’s a bubble but right now I’m making bank, so…”. Finally, after those investors decide it’s been a good run, they cash out and the bubble truly starts bursting.
So right now everyone knows it’s a bubble. What we’re seeing is the big investors trying to squeeze every last billion out of it.
This time they wasted so much money that they’re trying to foist the bad investments on retail investors with overblown valuations and IPOs before cashing out.
You seem to be talking about SpaceX, but I gotta point out that with the fast-track it’s not just the retail investors holding the bag. It’s anyone with a retirement fund.
I was thinking retail investors — whether individually or through an index fund — covered both. If you have any control over your retirement funds though, you can use that control to get out of the stock market if you want. So it isn’t anyone with a retirement fund, just nearly anyone with a retirement fund.
He didn’t make it into the s&p 500. NASDAQ bent over and opened wide though.
Yes.
Yeah, heard it all before, and I’m very familiar with the structural “curiosities” of the existing investment landscape.
Very few people correctly called the problems with 2007-2008. Not none, but few. And with soooooo many people mindlessly on the “it’s a bubble!” bandwagon so early, a lot of accuracy and legitimacy is lost months or years beforehand for no other reason than why conspiracy theory people say “we’ll get UFO disclosure this year!” Or “This year the Cubs/Arsenal/Red Sox will do it!” It’s just the thing they say until one time they’re right.
I’m not telling you it won’t happen in a sense… But it’s not going to happen how or when you think. IMO, you’re looking at a partial stuttering effect maaaaaybe late winter like Q1 2027, and that’s about it. There’s to much alternate demand for everything LLM companies are already buying up to create a full and similar bubble like the Dot Com bubble.
I mean yeah sure, if you choose to ignore all of the evidence.
https://finance.yahoo.com/news/very-troubling-ais-self-investment-spree-sets-off-bubble-alarms-on-wall-street-160524518.html
https://www.nbcnews.com/business/economy/openai-nvidia-amd-deals-risks-rcna234806
https://www.cnn.com/2025/10/07/business/openai-nvidia-bubble-nightcap
https://www.forbes.com/sites/greatspeculations/2025/11/28/how-the-10-trillion-ai-bubble-pops/
Specifically, the circular nature of the investment and funding is hard to brush off and ignore.
You’re only cherry-picking some panic-mongering bias-confirming reporting here. Honestly - Yahoo? NBC? CNN? Why not Grandma’s Facebook shares and the Babylon Bee?
Plus, did you even read the Forbes article? The opening paragraph is literally what I’m trying to tell you.
For example, one huge thing you’re missing this is that the large companies of the Big 4 with frontier models aren’t taking out loans for this. It’s all the startups that are trying to draft on them that are using loans. THAT is a danger, that tens of thousands of idiots will sink a corner of the financial system from the sheer weight of defaulting loans because ChatGPT said their shit sandwich idea was “revelatory, and honestly, a great idea!”
But that’s not going to tank OpenAI or Google. Google has cash money. OpenAI as a fair bit of cash money. Meta has cash money. Anthropic had cash and then Trump tried to sink them… but they might be OK in the long run. They’re doing great on the code side of things.
What you’re also erroneously assuming points to a bubble points to how OpenAI, Google, Whatever Musk calls his stupid company this week, and Palantir all are doing to avoid classic bubble economics of huge loans to pay back. You don’t pay back stocks. It’s equity. That’s cash trading hands, not loans that come due one day. Hell, Antropic is buying server time from Grok - that’s real cash trading hands to perform a service. That’s not a bubble, that’s the kind of thing that prevents a bubble. Like, bro, do you even bubble?
I don’t claim I know when the correction will happen but I wonder what massive alternative demand you see for the mountains of highly specialised server gpus in storage that will be obsolete in maybe 3 years time? For many of them that means likely before they will ever be turned on. The dotcom bubble created infrastructure that was, largely, not obsolete when the bubble bursted and made a lot of sense to salvage. That is a fundamental difference to inference infrastructure.
The WorldCom fiber layouts were akin to the railroad bubble in the 1840s, in that those were pathways with nothing to use them. I can see the parallels here, but the difference is GPUs aren’t nailed to the ground. They can be moved to demand, unlike railway lines and fiber lines.
GPUs process data. They don’t spoil or expire. Sure, they’ll lose value, but it’s not like they stop being useful, even if highly specialized. Hell, even selling them second hand to China with an export waiver would be a way to recoup value. So already, the premise is flawed in that, specialized or not, China will use them. Or the EU or universities looking for a deal and building out their own local processing.
Both rail and communication infrastructure lead to some useless connections but much of it was no useless, in both cases. GPUs are not bolted to the ground but they do become obsolete no matter if you deny it or not. The issue is that the real costs is in using GPUs is very different from these previous bubbles. Those obsolete GPUs will cause much higher operating costs than newer generations, to the point where they won’t be interesting to use even if you gave them away for free. To make matters worse, other infrastructure is much more flexible in its use, one can transport all sorts of things on railways, one can send all sorts of data on communication infrastructure. Those specialised GPUs aren’t very useful for anything other than a fairly narrow use case.
I think you do not fully appreciate the crazy amounts of GPUs we are talking about here. China has no massive real shortage of GPUs. They managed to get black market GPUs more or less directly from Nvidia just fine. Nor are European universities IT wastelands without compute capabilities. But even if they’d go crazy on expandig compute infrastructure with outdated power hungry GPUs, that would be barely more than a drop in the ocean. Nvidia does have to resort to circular financing to keep the boom cycle accelerating, with GPUs going just to some storage facility if they exist at all. That is not how healthy demand looks like.
In a $10 Trillion bubble, what percentage, exactly, are the GPUs you’re taking about?
Is it 10%? Are there a Trillion dollars worth of GPUs you’re worried about? Or less?
If you are talking about GPUs being only a small share of the overall total sum, bad news that the supporting infrastructure is also to a large extend tailor made for that very narrow use case. No one else will need such huge data center facilities designed specifically for GPUs, that includes also the non GPU components. And the infrastructure is the only thing of substance of this bubble. The models aren’t it. Open weight models are on the heels of the closed models. As soon as they are good enough for common applications, the business case for charging billions is slowly evaporating.
You are also mistaken, I am not worried about GPUs. I am merely stating that they and their server infrastructure (which is tailor made for them) are rapidly getting obsolete equipment by their nature and while the clock is ticking they are largely not even being used. This is fundamentally different from the dotcom and railway bubble.