I have not been paying much attention to the whole circular deal thing that NVIDIA is supposedly doing. As in, they invest in their clients, who buy their products.
Can anyone who actually understands finance please explain a couple things to me?
1. Are those accusations are true in a significant way, and are actually a bad thing?
2. This claimed $673B in sales, how much of it comes from NVIDIA's own money, invested into their clients? Is there any way to know that?
What do you mean by “if true”? It’s a fact. It’s “only” bad if what they’re investing in goes south, because NVIDIA gets hit twice: it loses money on the investment and loses the GPU demand.
NVIDIA says it has invested nearly $50B in frontier labs. According to NVIDIA, “the AI labs for which NVIDIA expects to leverage its balance sheet should account for roughly one-quarter of NVIDIA’s business next year.”
To be clear, this doesn’t mean 1/4 of $673B is NVIDIA money.
I didn't take enough time to be clear. What I really meant was something like: are they giving these companies money to directly buy their own products, or to spend on other things, so that they can grow enough to be able to buy NVIDIA products?
The distinction does make a difference and it is the former.
These circular deals are two paired transactions:
1. Nvidia buys equity in an AI lab or cloud provider with cash.
2. The counterparty agrees to buy X number of GPUs from Nvidia and in exchange Nvidia guarantees to rent some Y fraction of the compute if the counterparty cannot find customers.
This structure goes south during a pullback because all this liquidity Nvidia is essentially providing vanishes and contracts rapidly if the counterparty cannot find customers.
The other circular deal type is via private equity and the Special Purpose Vehicle (SPV).
1. The private equity firm loans money to the SPV.
2. The SPV buys GPUs from Nvidia for a data center.
3. Nvidia guarantees to the private equity firm residual value of the GPU.
This deal also breaks down if the demand for GPU compute never materializes because now Nvidia is on the hook to the private equity firm (the lender) for the residual value of the GPU, which again saps Nvidia's liquidity.
Basically these deals are extremely sharp double edged swords. As long as demand for compute outpaces the compute capacity Nvidia can provide, Nvidia's revenues grow exponentially. But if demand growth slows, stops, or goes negative, Nvidia is suddenly on the hook for their counterparties' losses. Suddenly Nvidia's cash flow goes extremely negative and the company's financial situation becomes dicey.
I understand it is the former. They basically give money to be spent on compute, meaning it goes to hyperscalers who themselves buy NVIDIA GPUs. That’s how you end up with OpenAI and Anthropic together representing more than 70% of the hyperscalers AI revenue, and >40% of the overall Google cloud revenue.
Another thing NVIDIA does: when hyperscalers are looking for debt to build more datacenter capacity, NVIDIA offers to be a backstop in case the compute isn’t actually used. If we take CoreWeave for example, NVIDIA has ownership in it, and also sell them GPUs, and also goes to the banks telling them they will for sure buy the unused capacity as a way to reduce the bank risks.
If I give you 10 dollars, and then you put it in your pocket, and then you take it out again and give me 10 dollars back - no actual economic growth occurred. It's simply shuffling money around; the amount stays the same.
Two economists are walking down a forest path and they come across a piece of animal excrement. The first economist turns to the second and says "I'll give you $100 to eat that!" The second one eats it and the first one gives him $100. They start walking again and a few minutes later they come to another piece of animal excrement. The second economist now looks at the first and says "I'll give you $100 to eat!" The first one eats it and collects his money.
They walk a bit more and the first one says, "You know, I gave you $100 to eat shit, and you gave me the same $100 to eat shit. I can't help but think we both just ate shit for nothing." The second one responds, "That's not true at all! We increased the GDP by $200!"
But it’s not just dollars changing hands. If I invest N dollars in your hot-dog business, and you use some of that capital to buy equipment from me, I get revenue from the sale, and I still own an investment in your business. If you succeed, that investment can also make me money.
If we assume they've invested up to $70B in other companies, which is the estimated value of their equity investments, then that implies that a maximum of 10% of that estimated revenue demand is coming directly circularly... and that assumes these companies spend the entirety of their invested capital on Nvidia infr in one year which seems unlikely so probably much lower.
Still that's not to say these companies aren't leveraging the Nvidia capital with others' in a way that magnifies or multiplies some of the effect.
But it looks like a second order contributor unless Nvidia's actions are acting like a backstop that causes way more risk and leverage to pile up in a way that could come tumbling down
This would require data centre cap-ex north of 1T and revenues from non AI companies in the same order of magnitude. Is it realistic to scale up data centre roll-outs? Are regular companies ready to re-allocate 1T? And this all happens in an environment where rates go up and many of the companies are not profitable?
> Small models are rapidly growing in capability, require less compute to train and serve
According to Jevons' paradox a reduction in resource requirements (improved resource efficiency for the same payoff) leads to an increase in demand. This stops working when demand for compute is completely exhausted, but we are very far from that. There's even some very silly predictions floating around (see the latest Dwarkesh Patel podcast) that say compute will soon be most of the economy, even dictating market interest rates. Now, that has to be wrong, but the directional outlook is closer to correct than "very small and efficient models mean there will be ~0 demand for HPC-like compute".
I don’t think this line of thinking is particularly robust because it ignores how AI is being used and where the resource usage is coming from.
Right now there are a small number of very very resource intensive use cases that are being subsidized by OpenAI and Anthropic. There are people generating millions of lines of code because it’s basically free at the point of use, despite the code producing very little value. Anthropic and OpenAI frequently “reset” customer limits to allow them to use even more resources at no additional cost.
The majority of use cases across business are not generating millions of lines of code per employee. The majority of businesses need just a little bit of automation to radically improve the way they operate. A software engineer making endless projects because it’s free to do so might use hundreds of billions of tokens per year, but an entire manufacturing business could be revolutionized with a few million tokens per year.
I think 2 things can be true:
1. There is very little penetration of AI across the economy and huge room to grow in the number of businesses deriving economic value from AI
2. The compute usage today is vastly overrepresented by usage outliers who are not paying the cost of their usage and will stop when forced to pay the cost
We could see AI usage 10x while seeing compute decrease 10x if the type of usage shifts. Most businesses just need smarter macros.
> Anthropic and OpenAI frequently “reset” customer limits to allow them to use even more resources at no additional cost.
Surely this applies to fixed-price subscriptions, not per-token spend? Large enterprises (the "very very resource intensive" large-scale users) have to pay per token.
A lot of companies avoid paying for usage by encouraging their employees to use individual subscriptions. Outside of the short lived tokenmaxxing fever dream, enterprises are conscious of their usage with companies like Uber and Amazon reigning in their usage massively and companies like Ramp building their own routers for cost minimization.
Facebook is reportedly the company that spent $500 million in a single month on tokens. There are individual non-enterprise users rotating multiple subscriptions incurring $10k+ in tokens per subscription. Facebook’s $500 million month… is equivalent to ~10k individual subscriptions which could be as little as a few thousand of the heaviest users. That’s $500 million when billed on usage, or ~$2 million on plans.
The reason resets are such a big deal (people have set up websites to track them, tweets announcing them get millions of impressions) is because there are huge numbers of users pushing their plan limits every single day. If there was huge demand from usage-based customers (the large enterprises) that OpenAI and Anthropic couldn’t meet, they wouldn’t be handing out resets like candy.
I think a realistic belief is that Anthropic and OpenAI have vastly overstated demand and are using resets as a way to keep usage artificially inflated at a substantial financial cost. I’d guess fixed price plan users make up at least 95% of usage.
Jevons paradox is a veridical paradox, which, as you said, means that it's a true statement that merely looks wrong because it is counterintuitive.
I know that some people think that the word "paradox" should be only used to refer to antinomy paradoxes which are often called "true paradoxes" (such as "this sentence is false") which lead to a contradiction without requiring a flaw in reasoning.
Yeah, I think there's a tendency to underestimate how much demand is still gated behind cost constraints. The market for this is HUGE.
The PC era, call it 1975-2005, was one of the greatest wealth creation events in history, was characterized by the cost of the underlying commodity dropping mercilessly for the whole time. Each time it did, the space of problem you could solve with a PC would increase, to the point that by the end, they were both replacing mainframes and powering users who do nothing but chat and post cat pictures.
Could there be a correction in the short run? Quite possibly. I think an underestimated last mile problem is just the massive weight of bureaucracy and human process inertia. But in the long run, cheap, efficient intelligence is a new engineering capability that we've just begun to even explore.
> how much demand is still gated behind cost constraints. The market for this is HUGE.
I think this misses the actual limits here.
The problem isn't demand it's, "how much people are willing to spend on it".
Cheap AI has to be served on cheap compute, and if inference gets cheap enough to unlock massive usage numbers, by definition it also doesn't require anywhere near as much infrastructure per unit of demand.
Take DeepSeek serving ~100T tokens/day, depending on workload and utilization, you're potentially talking about only a few thousand last-gen GPUs. With current-gen GPUs maybe closer to ~1,000, and with Rubin even fewer I will be damned if I could get my hands on one.
That's the part I think people are missing when they extrapolate token demand into enormous infrastructure or AI revenue.
Yes usage will explode. But if the cost per unit collapses, the revenue doesn't necessarily go up with it.
You can't simultaneously argue that intelligence becomes so cheap that everyone uses enormous amounts of it, while also assuming customers will somehow spend trillions of dollars a year consuming it.
There is no obvious $1T customer-facing AI revenue number at the end of this rainbow in the short/medium term.
The average person isn't going to spend anything remotely comparable to what they spend on a car every year for an AI service. Even businesses have budgets now, huge demand doesn't matter if the willingness to pay isn't there.
The only path I can see to numbers like that is AI consuming existing business domains, even then it's very thin.
Say SaaS + legal + consulting + BPO + various other service industries collectively represent something like $10-20T globally.
Even if AI eventually replaces an enormous portion of that, it's probably not doing so at the same price. Why would customers switch otherwise?
Either the AI product has to be dramatically better—which is difficult for mature workflows—or dramatically cheaper, which is much more plausible.
If it replaces $10-20T of existing services at roughly 1/10th or 1/100th (more likely) the cost, then you're looking at maybe a ~$1T AI revenue opportunity after replacing an absurdly large fraction of the existing service economy.
Who are now unemployed and can't pay for shit.
And that's before competition.
I think it's crazy to assume AI companies won't compete aggressively on price. As capabilities diffuse, smaller models catch up, inference hits pareto frontier the open-source alternatives have already improved and caught up, margins on routine intelligence should compress "hard" (emphasis on "hard").
We've already seen how difficult adoption can be even when the technology looks impressive on paper. Cheap here means 100x cheaper for 10x more demand that's a net 10x loss before any software or hardware optimizations.
So yes, I completely agree that cheap intelligence can bring an enormous amount of new usage.
"I just don't think usage means revenue." (you can plaster it on a wall if you want to, "usage doesn't mean revenue", if you want to find that out I have foss software bridge to sell)
The PC analogy actually reinforces this if you really think about it.
Compute became "vastly more useful" while the cost per unit of compute collapsed. Society captured enormous value, but all computer companies are literal failing giants without the AI hype. Value got caught by people who provided productionization.
Now if people expect AI to self productize itself I am happy to tell your try it. We all saw how OpenAI fell behind Anthropic because they thought that would work...
Google couldn't productize the search, instead they sold the eye balls and web-real-estate. Maybe that's the AI business model, but that's not $1T worth given you need to unglue people from other stuff.
Unless we get something approaching genuine ASI producing so much additional economic value that entirely new trillions, I don't see a path to $1-2T in direct AI revenue from customers.
The market simply can't absorb that level of spending.
Demand can be effectively infinite at the right price. But I think people are delusional on HN and SF if they think that number is in Trillions like the investments seem to suggest.
I am not saying Nvidia will fall tomorrow but someone will have to pull the breaks before this car goes to hell.
If AI compute is a transformative technology compared to industrialization (that's a huge "if", essentially positing a singularity-like outcome), that $1T-$2T/yr at current prices might be a tiny fraction of future GDP (real incomes), thus actually quite sustainable.
"The problem isn't demand it's, "how much people are willing to spend on it".
Lol its not even that - its what can I do with it? Which eventually has to show up somehow in the financials - from a macroeconomic stand point. Software production is microeconomic.
Every quarter I see a similar analysis, similar projection. Yet, they keep posting these insane numbers. Everyone knows it’s a bubble, the problem is determining the top. Nvidia is continuously showing the top is far far higher than everyone imagines.
I don't get why people say Nvidia is a bubble. They are selling products now, not in the future! If AI market collapses (I doubt it will happen), they will still be selling GPUs. They will make less money, but that is expected
the bubble doesnt mean they are worthless only that after a pop the value of stock will drop significantly. it is out of their control if people overvalue the stock, and thats what creates the bubble which will eventually need to pop for self correction - but might trigger a massive oversale bringing the stock below actual value and causing all sorts of problems that will challenge the solvency of the company (basically challenging their liquid funds vs how much debt that they backed to their stock value). if they survive that then a bounce back is expected and buying while they were low would get you profit again. if they overleveraged themselves during the bubble bc they bought into the hype themselves, then they could face serious financial troubles and be susceptible to getting bought out.
But isn't that supposed to happen when you create a hit product? I imagine some people said similar things about Apple when iPhone started selling like water, but I don't think that it was the majority like here
Bubble means inflated not fake, i.e when they make less money their stock will crash and bubble will pop, which is what people buying the stock today are concerned with , is this going to hold
>Small models are rapidly growing in capability, require less compute to train and serve
Must be very clear that China’s undercut strategy, which is a well-known and studied tactic that they’ve used for a long time, it is absolutely dominating this point.
Right now you can LLM, code, make songs, images, and esp video on gaming hardware in your PC that would’ve been absolutely datacenter shit last year.
So the question will be does the scaling continue to benefit efficiency or ability?
If ability (needs datacenter storage and performance), how much better can the code get? How much more realistic in the images videos get? There are definitely strides to be made everywhere, but man, just like the bottleneck wasn’t coding, I’m not sure the creation bottleneck is rendering.
I couldn’t find the 1000b number, but if that is “ai buildout capex” that means you think that 67% of every dollar spent on building data centers, training models, and running inference, is going directly to nvidia.
Which is just silly on the face of it. Data centers need concrete, copper, DRAM, SSDs, and labour. That alone will cost more than 33%.
Yeah but remember 1) stuff gets more expensive the more it's "processed"/further up the chain. It's hard to think of something further up the value chain than a modern gpu server. Just look at one input, asml machines. 2) a lot of dram, labor etc is also baked into the gpus.
Their profits are incredible too, they have 62% net margin! Last quarter $60b profit on $96b revenue. Since the "AI bubble going to pop" terrible takes 2 years ago, NVDA has made $300b in profits.
There's a specific type of person who makes a repeated joke like "sell shovels in a gold rush" - they think they have made the most insightful comment possible.
SK Hynix (together with Samsung and Nvidia) claims $700B investment [1], Samsung itself is investing $70B, Micron $25B, Sandisk announced $31B today [2]
Margin is insane. They made roughly double net income, pure profit, what Apple did (even if you take out the ~8B in paper gains from their investments in other AI shops) on 13B less revenue.
As someone who uses AI all day it all makes sense. However, if AI is going to have serious impact white collar jobs as some people predict, the demand will decline. People without jobs won't pay for expensive subscriptions or API prices and the economy will be in recession.
For me the important question is where the economy will be in the next 5 years. Because if the economy is doing well, I have no doubt the AI demand will continue to sky rocket. I don't think it matters to Nvidia how uses their compute, closed or open models. The win either way.
The financial press regurgitates it and in return they get ad spend and exclusive stories.
It’s mutual dependency unfortunately, and will remain that way whilst there are shareholders and investors who own the publications and seek only ongoing returns
Do not short it. I repeat, do not do this! Do not play the game of capitalism against expert capitalists who have spent the last few decades rigging the game more an more in their own favor. In very rare cases may be lucky but only for a short while. The more trades you make the more the odds stack against you, and you will loose everything.
In a game where the odds are stacked against you, the only winning move is not to play.
> The more trades you make the more the odds stack against you, and you will loose everything.
In general this is true. Participation is not rigged with eg VTI/ITOT and VXUS/IXUS and a long enough time horizon.
There is no case where one cannot participate - doing nothing means inflation will eat away at assets.
I have a very brutish 50/50 international/US split (was 30% international before 2024 when Trump promised to destroy the US economy and started to act on that...). Each quarterly equity vest I put more in, and realize capital "losses" while buying the near equivalent security when there is an opportunity to do so.
It’s a shitty advice (the one about not playing, not the one about not buying shorts). The winning move is to have diversified longs with long horizon. That’s all it takes to participate in profits from “rigged” system.
I disagree. The winning move is to organize, and stand as workers in solidarity against the owning classes, engage in direct action (including boycotts and strikes) until the ruling class changes the rules in our favor, or better yet, abolishes this godforsaken economic system of capitalism in favor of a system that rewards workers and not our exploiters.
Right answer, but the anti-captialist reasoning is rubbish. The simple reason why you do not want to short a stock is that it can go up infinitely, and nVidia has a four year history of going up. There is also good reason to think nVidia will hit the numbers, both because AI is booming and because nVidia is conservative in its forcasts.
No capitalist conspiracy theory is needed, save the conspiracy theories for the Trumpers.
TBF there are absolutely horrible platforms like Robinhood that aren't far off what the above poster alluded to. At one point payment for order flow was the majority of their revenue or near it!
The boring staid firms like Fidelity/Vanguard have vastly superior products not designed to incentivize gambling
It's all factored in! The unimaginably vastness of the cosmos: Past, future, at astronomical through Planck scales. You have no agency; free will is an emergent phenomenon. You are not smarter than all those quants with their Ivy league degrees and 7 figure salaries, surely; the only winning move is not to play.
I have not been paying much attention to the whole circular deal thing that NVIDIA is supposedly doing. As in, they invest in their clients, who buy their products.
Can anyone who actually understands finance please explain a couple things to me?
1. Are those accusations are true in a significant way, and are actually a bad thing?
2. This claimed $673B in sales, how much of it comes from NVIDIA's own money, invested into their clients? Is there any way to know that?
What do you mean by “if true”? It’s a fact. It’s “only” bad if what they’re investing in goes south, because NVIDIA gets hit twice: it loses money on the investment and loses the GPU demand.
NVIDIA says it has invested nearly $50B in frontier labs. According to NVIDIA, “the AI labs for which NVIDIA expects to leverage its balance sheet should account for roughly one-quarter of NVIDIA’s business next year.”
To be clear, this doesn’t mean 1/4 of $673B is NVIDIA money.
I didn't take enough time to be clear. What I really meant was something like: are they giving these companies money to directly buy their own products, or to spend on other things, so that they can grow enough to be able to buy NVIDIA products?
The latter sounds a lot more reasonable to me.
The distinction does make a difference and it is the former.
These circular deals are two paired transactions: 1. Nvidia buys equity in an AI lab or cloud provider with cash. 2. The counterparty agrees to buy X number of GPUs from Nvidia and in exchange Nvidia guarantees to rent some Y fraction of the compute if the counterparty cannot find customers.
This structure goes south during a pullback because all this liquidity Nvidia is essentially providing vanishes and contracts rapidly if the counterparty cannot find customers.
The other circular deal type is via private equity and the Special Purpose Vehicle (SPV). 1. The private equity firm loans money to the SPV. 2. The SPV buys GPUs from Nvidia for a data center. 3. Nvidia guarantees to the private equity firm residual value of the GPU.
This deal also breaks down if the demand for GPU compute never materializes because now Nvidia is on the hook to the private equity firm (the lender) for the residual value of the GPU, which again saps Nvidia's liquidity.
Basically these deals are extremely sharp double edged swords. As long as demand for compute outpaces the compute capacity Nvidia can provide, Nvidia's revenues grow exponentially. But if demand growth slows, stops, or goes negative, Nvidia is suddenly on the hook for their counterparties' losses. Suddenly Nvidia's cash flow goes extremely negative and the company's financial situation becomes dicey.
I understand it is the former. They basically give money to be spent on compute, meaning it goes to hyperscalers who themselves buy NVIDIA GPUs. That’s how you end up with OpenAI and Anthropic together representing more than 70% of the hyperscalers AI revenue, and >40% of the overall Google cloud revenue.
Another thing NVIDIA does: when hyperscalers are looking for debt to build more datacenter capacity, NVIDIA offers to be a backstop in case the compute isn’t actually used. If we take CoreWeave for example, NVIDIA has ownership in it, and also sell them GPUs, and also goes to the banks telling them they will for sure buy the unused capacity as a way to reduce the bank risks.
So you can add that to the whole circular thing
If I give you 10 dollars, and then you put it in your pocket, and then you take it out again and give me 10 dollars back - no actual economic growth occurred. It's simply shuffling money around; the amount stays the same.
Two economists are walking down a forest path and they come across a piece of animal excrement. The first economist turns to the second and says "I'll give you $100 to eat that!" The second one eats it and the first one gives him $100. They start walking again and a few minutes later they come to another piece of animal excrement. The second economist now looks at the first and says "I'll give you $100 to eat!" The first one eats it and collects his money.
They walk a bit more and the first one says, "You know, I gave you $100 to eat shit, and you gave me the same $100 to eat shit. I can't help but think we both just ate shit for nothing." The second one responds, "That's not true at all! We increased the GDP by $200!"
This is basically true for any entertainment.
But it’s not just dollars changing hands. If I invest N dollars in your hot-dog business, and you use some of that capital to buy equipment from me, I get revenue from the sale, and I still own an investment in your business. If you succeed, that investment can also make me money.
Wrong framing.
Nvidia is investing - it is using its shareholder's cash under the assumption it will create value for them.
If we assume they've invested up to $70B in other companies, which is the estimated value of their equity investments, then that implies that a maximum of 10% of that estimated revenue demand is coming directly circularly... and that assumes these companies spend the entirety of their invested capital on Nvidia infr in one year which seems unlikely so probably much lower.
Still that's not to say these companies aren't leveraging the Nvidia capital with others' in a way that magnifies or multiplies some of the effect.
But it looks like a second order contributor unless Nvidia's actions are acting like a backstop that causes way more risk and leverage to pile up in a way that could come tumbling down
This would require data centre cap-ex north of 1T and revenues from non AI companies in the same order of magnitude. Is it realistic to scale up data centre roll-outs? Are regular companies ready to re-allocate 1T? And this all happens in an environment where rates go up and many of the companies are not profitable?
Sold after the price rise from this announcement. They can make the sales projection, but it doesn't mean they'll hit it:
1. Small models are rapidly growing in capability, require less compute to train and serve
2. There are more suppliers now, both in China & the US (OpenAI even have their own inferencing hardware now)
3. Memory still constrains how much they can ship in the short term
> Small models are rapidly growing in capability, require less compute to train and serve
According to Jevons' paradox a reduction in resource requirements (improved resource efficiency for the same payoff) leads to an increase in demand. This stops working when demand for compute is completely exhausted, but we are very far from that. There's even some very silly predictions floating around (see the latest Dwarkesh Patel podcast) that say compute will soon be most of the economy, even dictating market interest rates. Now, that has to be wrong, but the directional outlook is closer to correct than "very small and efficient models mean there will be ~0 demand for HPC-like compute".
I don’t think this line of thinking is particularly robust because it ignores how AI is being used and where the resource usage is coming from.
Right now there are a small number of very very resource intensive use cases that are being subsidized by OpenAI and Anthropic. There are people generating millions of lines of code because it’s basically free at the point of use, despite the code producing very little value. Anthropic and OpenAI frequently “reset” customer limits to allow them to use even more resources at no additional cost.
The majority of use cases across business are not generating millions of lines of code per employee. The majority of businesses need just a little bit of automation to radically improve the way they operate. A software engineer making endless projects because it’s free to do so might use hundreds of billions of tokens per year, but an entire manufacturing business could be revolutionized with a few million tokens per year.
I think 2 things can be true:
1. There is very little penetration of AI across the economy and huge room to grow in the number of businesses deriving economic value from AI
2. The compute usage today is vastly overrepresented by usage outliers who are not paying the cost of their usage and will stop when forced to pay the cost
We could see AI usage 10x while seeing compute decrease 10x if the type of usage shifts. Most businesses just need smarter macros.
> Anthropic and OpenAI frequently “reset” customer limits to allow them to use even more resources at no additional cost.
Surely this applies to fixed-price subscriptions, not per-token spend? Large enterprises (the "very very resource intensive" large-scale users) have to pay per token.
A lot of companies avoid paying for usage by encouraging their employees to use individual subscriptions. Outside of the short lived tokenmaxxing fever dream, enterprises are conscious of their usage with companies like Uber and Amazon reigning in their usage massively and companies like Ramp building their own routers for cost minimization.
Facebook is reportedly the company that spent $500 million in a single month on tokens. There are individual non-enterprise users rotating multiple subscriptions incurring $10k+ in tokens per subscription. Facebook’s $500 million month… is equivalent to ~10k individual subscriptions which could be as little as a few thousand of the heaviest users. That’s $500 million when billed on usage, or ~$2 million on plans.
The reason resets are such a big deal (people have set up websites to track them, tweets announcing them get millions of impressions) is because there are huge numbers of users pushing their plan limits every single day. If there was huge demand from usage-based customers (the large enterprises) that OpenAI and Anthropic couldn’t meet, they wouldn’t be handing out resets like candy.
I think a realistic belief is that Anthropic and OpenAI have vastly overstated demand and are using resets as a way to keep usage artificially inflated at a substantial financial cost. I’d guess fixed price plan users make up at least 95% of usage.
There is no paradox, simply (a/b) increasing tells you nothing about a nor b. Jevons only “destroys” the (wrong) intuition that total b would decrease
"paradox" is an overloaded term.
Jevons paradox is a veridical paradox, which, as you said, means that it's a true statement that merely looks wrong because it is counterintuitive.
I know that some people think that the word "paradox" should be only used to refer to antinomy paradoxes which are often called "true paradoxes" (such as "this sentence is false") which lead to a contradiction without requiring a flaw in reasoning.
Thank you for a great comment; I learned two new words today!
Yeah, I think there's a tendency to underestimate how much demand is still gated behind cost constraints. The market for this is HUGE.
The PC era, call it 1975-2005, was one of the greatest wealth creation events in history, was characterized by the cost of the underlying commodity dropping mercilessly for the whole time. Each time it did, the space of problem you could solve with a PC would increase, to the point that by the end, they were both replacing mainframes and powering users who do nothing but chat and post cat pictures.
Could there be a correction in the short run? Quite possibly. I think an underestimated last mile problem is just the massive weight of bureaucracy and human process inertia. But in the long run, cheap, efficient intelligence is a new engineering capability that we've just begun to even explore.
"Yeah, I think there's a tendency to underestimate how much demand is still gated behind cost constraints. The market for this is HUGE."
This is just hyperbolic nonsense.
There is a desire from a certain group of people of make-believe - doesn't mean the 'demand' is actually real given the economics.
> The market for this is HUGE.
Source(s)?
The backlog of every software team on the planet being anywhere between 1 and 100 years long at human burn rates.
> how much demand is still gated behind cost constraints. The market for this is HUGE.
I think this misses the actual limits here.
The problem isn't demand it's, "how much people are willing to spend on it".
Cheap AI has to be served on cheap compute, and if inference gets cheap enough to unlock massive usage numbers, by definition it also doesn't require anywhere near as much infrastructure per unit of demand.
Take DeepSeek serving ~100T tokens/day, depending on workload and utilization, you're potentially talking about only a few thousand last-gen GPUs. With current-gen GPUs maybe closer to ~1,000, and with Rubin even fewer I will be damned if I could get my hands on one.
That's the part I think people are missing when they extrapolate token demand into enormous infrastructure or AI revenue.
Yes usage will explode. But if the cost per unit collapses, the revenue doesn't necessarily go up with it.
You can't simultaneously argue that intelligence becomes so cheap that everyone uses enormous amounts of it, while also assuming customers will somehow spend trillions of dollars a year consuming it.
There is no obvious $1T customer-facing AI revenue number at the end of this rainbow in the short/medium term.
The average person isn't going to spend anything remotely comparable to what they spend on a car every year for an AI service. Even businesses have budgets now, huge demand doesn't matter if the willingness to pay isn't there.
The only path I can see to numbers like that is AI consuming existing business domains, even then it's very thin.
Say SaaS + legal + consulting + BPO + various other service industries collectively represent something like $10-20T globally.
Even if AI eventually replaces an enormous portion of that, it's probably not doing so at the same price. Why would customers switch otherwise?
Either the AI product has to be dramatically better—which is difficult for mature workflows—or dramatically cheaper, which is much more plausible.
If it replaces $10-20T of existing services at roughly 1/10th or 1/100th (more likely) the cost, then you're looking at maybe a ~$1T AI revenue opportunity after replacing an absurdly large fraction of the existing service economy.
Who are now unemployed and can't pay for shit.
And that's before competition.
I think it's crazy to assume AI companies won't compete aggressively on price. As capabilities diffuse, smaller models catch up, inference hits pareto frontier the open-source alternatives have already improved and caught up, margins on routine intelligence should compress "hard" (emphasis on "hard").
We've already seen how difficult adoption can be even when the technology looks impressive on paper. Cheap here means 100x cheaper for 10x more demand that's a net 10x loss before any software or hardware optimizations.
So yes, I completely agree that cheap intelligence can bring an enormous amount of new usage.
"I just don't think usage means revenue." (you can plaster it on a wall if you want to, "usage doesn't mean revenue", if you want to find that out I have foss software bridge to sell)
The PC analogy actually reinforces this if you really think about it. Compute became "vastly more useful" while the cost per unit of compute collapsed. Society captured enormous value, but all computer companies are literal failing giants without the AI hype. Value got caught by people who provided productionization.
Now if people expect AI to self productize itself I am happy to tell your try it. We all saw how OpenAI fell behind Anthropic because they thought that would work...
Google couldn't productize the search, instead they sold the eye balls and web-real-estate. Maybe that's the AI business model, but that's not $1T worth given you need to unglue people from other stuff.
Unless we get something approaching genuine ASI producing so much additional economic value that entirely new trillions, I don't see a path to $1-2T in direct AI revenue from customers.
The market simply can't absorb that level of spending.
Demand can be effectively infinite at the right price. But I think people are delusional on HN and SF if they think that number is in Trillions like the investments seem to suggest.
I am not saying Nvidia will fall tomorrow but someone will have to pull the breaks before this car goes to hell.
If AI compute is a transformative technology compared to industrialization (that's a huge "if", essentially positing a singularity-like outcome), that $1T-$2T/yr at current prices might be a tiny fraction of future GDP (real incomes), thus actually quite sustainable.
"The problem isn't demand it's, "how much people are willing to spend on it".
Lol its not even that - its what can I do with it? Which eventually has to show up somehow in the financials - from a macroeconomic stand point. Software production is microeconomic.
Every quarter I see a similar analysis, similar projection. Yet, they keep posting these insane numbers. Everyone knows it’s a bubble, the problem is determining the top. Nvidia is continuously showing the top is far far higher than everyone imagines.
I don't get why people say Nvidia is a bubble. They are selling products now, not in the future! If AI market collapses (I doubt it will happen), they will still be selling GPUs. They will make less money, but that is expected
the bubble doesnt mean they are worthless only that after a pop the value of stock will drop significantly. it is out of their control if people overvalue the stock, and thats what creates the bubble which will eventually need to pop for self correction - but might trigger a massive oversale bringing the stock below actual value and causing all sorts of problems that will challenge the solvency of the company (basically challenging their liquid funds vs how much debt that they backed to their stock value). if they survive that then a bounce back is expected and buying while they were low would get you profit again. if they overleveraged themselves during the bubble bc they bought into the hype themselves, then they could face serious financial troubles and be susceptible to getting bought out.
But isn't that supposed to happen when you create a hit product? I imagine some people said similar things about Apple when iPhone started selling like water, but I don't think that it was the majority like here
Bubble means inflated not fake, i.e when they make less money their stock will crash and bubble will pop, which is what people buying the stock today are concerned with , is this going to hold
>Small models are rapidly growing in capability, require less compute to train and serve
Must be very clear that China’s undercut strategy, which is a well-known and studied tactic that they’ve used for a long time, it is absolutely dominating this point.
Right now you can LLM, code, make songs, images, and esp video on gaming hardware in your PC that would’ve been absolutely datacenter shit last year.
So the question will be does the scaling continue to benefit efficiency or ability?
If ability (needs datacenter storage and performance), how much better can the code get? How much more realistic in the images videos get? There are definitely strides to be made everywhere, but man, just like the bottleneck wasn’t coding, I’m not sure the creation bottleneck is rendering.
Posts that question the AI Endsieg are flagged now. That is supreme confidence in the numbers.
It’s crazy that their profit margin is above 50 %, before the AI craze I only knew such margins from drug trafficking cartels (supposedly).
Well, being best of the best always pays off.
It's obviously hard to understand to people bad at everything.
TSMC could charge more if they also lent the customer money to buy the chips
Same with ASMl who are the sole company behind their machines.
As could millions of other companies.
This is an unprecedented circular debt gamble. The market price is not based on affordability but distorted by the seller.
Or more accurately being top dog in a duopoly position pays off
And Apple
Hermes is close
The projected ai capex for this and next years is above 1000b/year. 673b/year doesn't sound so weird, if they manage to spent so much, obviously.
Capex covers a lot of non-gpu stuff, but yeah it doesn't sound crazy to me.
I couldn’t find the 1000b number, but if that is “ai buildout capex” that means you think that 67% of every dollar spent on building data centers, training models, and running inference, is going directly to nvidia.
Which is just silly on the face of it. Data centers need concrete, copper, DRAM, SSDs, and labour. That alone will cost more than 33%.
Yeah but remember 1) stuff gets more expensive the more it's "processed"/further up the chain. It's hard to think of something further up the value chain than a modern gpu server. Just look at one input, asml machines. 2) a lot of dram, labor etc is also baked into the gpus.
> Data centers need concrete, copper, DRAM, SSDs, and labour. That alone will cost more than 33%.
Surely the land is a double digit percentage of their budget? I know they build in the middle of nowhere, but even then.
A long time ago I put money into Qualcomm and then just dumped the remaining cash into Nvidia. That did well but wow do I wish I had done the reverse
Remember, Nvidia sells services too, not just GPUs. They also sell full racks of servers. This is just sales, not profits.
Their profits are incredible too, they have 62% net margin! Last quarter $60b profit on $96b revenue. Since the "AI bubble going to pop" terrible takes 2 years ago, NVDA has made $300b in profits.
A better question is what percent of that profit could be clawed back from loan guarantees. Unlike debt they dont give a consolidated number.
There's a specific type of person who makes a repeated joke like "sell shovels in a gold rush" - they think they have made the most insightful comment possible.
What if there are no memory chips for people to build hardware with, using nvidia components?
There's been lots of investment in memory
SK Hynix (together with Samsung and Nvidia) claims $700B investment [1], Samsung itself is investing $70B, Micron $25B, Sandisk announced $31B today [2]
[1] https://asia.nikkei.com/business/technology/artificial-intel... [2] https://www.wsj.com/tech/kioxia-sandisk-to-invest-more-than-...
Is this $673B coming mostly from BigAI?
Or from individuals and smaller companies?
Can they even find the production slots at TSMC?
Margin is insane. They made roughly double net income, pure profit, what Apple did (even if you take out the ~8B in paper gains from their investments in other AI shops) on 13B less revenue.
As someone who uses AI all day it all makes sense. However, if AI is going to have serious impact white collar jobs as some people predict, the demand will decline. People without jobs won't pay for expensive subscriptions or API prices and the economy will be in recession.
For me the important question is where the economy will be in the next 5 years. Because if the economy is doing well, I have no doubt the AI demand will continue to sky rocket. I don't think it matters to Nvidia how uses their compute, closed or open models. The win either way.
cool stuff!!!!
The way publications just regurgitate this stuff with no critical thought boggles my mind.
The financial press regurgitates it and in return they get ad spend and exclusive stories.
It’s mutual dependency unfortunately, and will remain that way whilst there are shareholders and investors who own the publications and seek only ongoing returns
Short it. NVDA is the most liquid market in history.
Do not short it. I repeat, do not do this! Do not play the game of capitalism against expert capitalists who have spent the last few decades rigging the game more an more in their own favor. In very rare cases may be lucky but only for a short while. The more trades you make the more the odds stack against you, and you will loose everything.
In a game where the odds are stacked against you, the only winning move is not to play.
This is a financial advice.
> The more trades you make the more the odds stack against you, and you will loose everything.
In general this is true. Participation is not rigged with eg VTI/ITOT and VXUS/IXUS and a long enough time horizon.
There is no case where one cannot participate - doing nothing means inflation will eat away at assets.
I have a very brutish 50/50 international/US split (was 30% international before 2024 when Trump promised to destroy the US economy and started to act on that...). Each quarterly equity vest I put more in, and realize capital "losses" while buying the near equivalent security when there is an opportunity to do so.
Not sure I have a choice here. My pension is already being gambled on it.
It’s a shitty advice (the one about not playing, not the one about not buying shorts). The winning move is to have diversified longs with long horizon. That’s all it takes to participate in profits from “rigged” system.
Or in general, short-term strategies don't work unless you've got the insider info
I don't know about that. I made a fair bit every time Musk opened his face hole.
And I made a lot off oil futures in high school thinking I had a plan, but it was just gambling
Well there's gambling and trading on sentiment, the latter of which I'm rather experienced in.
Yeah, to paraphrase an old quote “the best way to make a small fortune doing active trading is to start with a large fortune”.
I disagree. The winning move is to organize, and stand as workers in solidarity against the owning classes, engage in direct action (including boycotts and strikes) until the ruling class changes the rules in our favor, or better yet, abolishes this godforsaken economic system of capitalism in favor of a system that rewards workers and not our exploiters.
History clearly teaches us that it’s a losing move.
The winning move is to continually deliver value and realize when the market wants something else than what you have to offer.
Instead of blaming others, improve yourself and your station.
Life is too short to spend it putting up road blocks. Create a new, shorter circuit path instead.
The "marked" consolidated.
> Instead of blaming others, improve yourself and your station.
Yeah bro, just learn to code.
The market wants grifters and thieves. Can I take a course somewhere?
Oh no, WOPR/Joshua has gained LLM sentience.
(it is sage advice, but reminded me of nuclear war)
Right answer, but the anti-captialist reasoning is rubbish. The simple reason why you do not want to short a stock is that it can go up infinitely, and nVidia has a four year history of going up. There is also good reason to think nVidia will hit the numbers, both because AI is booming and because nVidia is conservative in its forcasts.
No capitalist conspiracy theory is needed, save the conspiracy theories for the Trumpers.
TBF there are absolutely horrible platforms like Robinhood that aren't far off what the above poster alluded to. At one point payment for order flow was the majority of their revenue or near it!
The boring staid firms like Fidelity/Vanguard have vastly superior products not designed to incentivize gambling
Care to post your shorts? lol
I'm long, I'm just looking for free money
“Markets can remain irrational longer than you can remain solvent”
It's all factored in! The unimaginably vastness of the cosmos: Past, future, at astronomical through Planck scales. You have no agency; free will is an emergent phenomenon. You are not smarter than all those quants with their Ivy league degrees and 7 figure salaries, surely; the only winning move is not to play.
You can play the game of life.
But most of us would be better off not gambling on high risk bets like shorting NVDA.