I don't think people realize just how screwed Anthropic is if their revenue doesn't grow at a crazy rate. They can be incredibly successful and still go bankrupt because of how much compute they are committed to buy. Everything has to go right, and if it doesn't then they go out of business.
What happens if Anthropic becomes a specialized high intelligence company because they can't afford the build out. Claude would be like an expensive Bloomberg Terminal or an investment letter that offers valuable and expensive intelligence.
I think about this because I pay $100/mo for Fable. How much more would I be willing to pay? I'm not sure yet. It depends on how smart the AI in the future becomes.
Being a specialized firm might work if they didn’t have $700 bn in commitments.
Already they have exclusive deals with trading firms who pay $$$ for faster models and more features. But there are only so many firms who will pay that much (and a lot of companies are pulling back on spend, not increasing it)
What does that imply about where SpaceX should have been? It feels like we are re running the WeWork IPO except without the investors who said "no way"
They didn't "invest". The Dems gave Intel free money to invest in fabs. The next President said the US public got ripped off by that and got a 10% stake "for free" to try and make up for it.
It was worth $8.9 billion when the US govt got it "for free". It's worth $23B today, even after the big drops recently. How is the US public worse off?
Not true. Is the Anthropic IPO happening this picosecond? At 50/50 odds, I'll bet on no. I'll bet on it for the next million picoseconds. How much are you willing to put on yes?
It may only be a short-term circumstance that you need a data centre to run AI for lucrative applications. Thirty years ago you needed a mainframe to beat Kaspersky at chess. Algorithm's got better, and now you can beat the mainframe that beat Kaspersky with a bot running on a desktop Pentium of the day[1]. You could still, today, devote a full hyperscale data centre to create a chess bot to beat all other chess bots, but nobody is doing that because there's no money in it.
The frontier is continually pushed outwards. At it's outer edge you need mainframes and data-centres. Once it pushes past, you don't. There wasn't a lot of money in Chess or protein folding, despite the later having huge potential for public benefit. There is a lot of money in coding and middle management, and that's what currently has tech-bros in a frenzy. However, there's no guarantee that what the frontier moves onto next will be lucrative. It might be more chess and protein folding. If, as Algorithm's and hardware improve, coding moves to local models (Sam Altman's nightmare) and nothing lucrative immediately replaces that, what are these companies going do do with their massive data-centres?
The scale of investment we're seeing these companies commit only makes sense if they have multiple decades to reap the returns. They may well not have that much time.
Chess is a fixed target, so as a benchmark, it can be saturated.
Reality can’t be saturated. The set of things you can accomplish with more intelligence is unbounded. So while cheap models will continue to proliferate, there will always be a massive market at the frontier, and that won’t ever change.
Intelligence is bounded by it's interface to reality. Until that problem is solved the set of what AI can accomplish in a reasonable timeframe is quite limited.
So is anyone with more than two brain cells to rub together. What does Anthropic have that MyNewAi Co doesn't? Open models are not bad, and anyone can pay google for GPU rent.
They do not have Claude. I use Pi and other harnesses, but it's not the same.
Microsoft has already proven that they do not have the software chops to deliver this. Even if they had the right product managers and UI experts, MS management would torpedo the project and ensure that it gets zero adoption.
And AWS has zero software capability in that regard.
And anyone with 3 can also see another trend. The hardware required for inference is rapidly falling. It's all following what happened with text-to-image image gen stuff almost to a t. It started out requiring super computing clusters, and now a days you can do high performance image gen on any plain old computer. I think within a few years we're going to be running frontier level models on a basic mid-range gaming hardware. You can already run quantized frontier level models on $1600 tier cards at high performance.
Seeing this bubble play out in real time is the most surreal thing. It's like watching a train wreck play out at a speed of a few inches per day. Like come on, just get off the track FFS.
Anthropic keeps nerfing their models to appease entrenched corporate overlords and it takes 5 seconds to switch to Deepseek or GLM. Can’t even use Claude to protect yourself against a cyber attack from a Chinese model based attacker for “safety”. What a joke.
Nerfed, proprietary models have have no future and these trillion dollar valuations are laughable.
So is anyone with even a minimally functioning brain. Anthropic has no viable business model because models are generic commodity. There's no lockin that Anthropic has because it's trivial for people to switch to a different provider. And whatever improvements Anthropic makes are quickly matched by other models within months.
So, it's a race to the bottom because they have to keep spending more and more money on training the models just to keep ahead of the competition, while prices are driven down by the availability of open models.
Chinese companies understand this, and that's why they give their models away in the open. The real money is going to be in integrations and custom services built on top of the models. The models themselves aren't what's valuable.
This is a real question, because I just don't know. Don't the models needs lots and lots of training that need the gobs of giant data centers? So you either need the infrastructure or the money to buy computing to make the models? I see that all the money is being invested in AI and data centers but I do not quite understand why.
The models are costly to make, but they're hard to monetize because there's not enough difference between them. It's a general commodity, so profit margins are thin because what companies can charge ends up being pushed down to the cost of inference.
> I see that all the money is being invested in AI and data centers but I do not quite understand why.
I am no economist or business person but AFAICT the massive amount of money being burned is because everyone in the space believes there will be a Google-like (for search) single winner who takes all when the dust settles on the AI race.
Because of this belief they are willing to burn any amount of money they can get their hands on to get to this mythical finish line as fast as they possibly can, seeing it as existential. Moving fast (particularly when your main cost is hardware capex) means moving expensively.
The real question is why anyone (let alone everyone) believes this single-winner theory in 2026 when all available evidence over the past few years shows that there is neither a moat nor a lock-in opportunity at all in this market.
For your product to be fit for purpose, it has to be able to automate your customer's ability to switch to a competing product with little to no friction. And all of the competing products are roughly the same in capabilities over a long enough timeline. Good luck building any kind of real moat with that burden.
But I guess after a certain point after you've taken on hundreds of billions of dollars of investment you're pot committed and just have to keep going through the motions even if you later see the futility, perhaps that's where the AI labs are now...? Either that or just common variety denial.
I feel like there is a case to be made that they will be the dominant players, but if we're talking about them charging a heavy premium and trying to target enterprise customers, then I don't know how well that'll go
they're already making a business model where enterprises buy tokens and put them in a pool for their users to use, and I feel like that's exactly where some companies will then compare that token pricing to open source models and find it hard to justify the prices, especially as it's quite obvious that many models https://news.ycombinator.com/formatdocare now "good enough"
> The real question is why anyone (let alone everyone) believes this single-winner theory in 2026
In the First World War, when they introduced the tank, everybody assumed its role would be to punch a big enough hole in the front lines that the horse cavalry could ride through and attack the enemy's supply lines, because That's How Wars End™. You fight the war you know, even when it's not the one you're actually fighting.
For your product to be fit for purpose, it has to be able to automate your customer's ability to switch to a competing product with little to no friction.
This is a really great identification of why, from an investing perspective, AI is nothing like past technologies.
There's a finite amount of training data available, and multiple companies have downloaded all of it. Arguably, trying to get a monopoly on training data is equivalent to getting monopoly access to the open internet + books.
That ain't going to happen, and if it somehow did in the US, things like Anna's Archive + libgen will save the rest of the planet from our self-imposed dark ages.
As for those data centers: The cost of inference and training on the same hardware is dropping much faster than Moore's Law. That multiplies with hardware advances, since Moore's Law isn't quite done yet.
So, those datacenters are depreciating at 50% every 18 months (Moore's law) at the same time as performance/$ and performance/watt of the software is dropping at some crazy rate like 10x per year. So, token per unit of data center is increasing by something like > 20x every 18 months, and that'll probably happen at least one more time. If it does, the datacenters will be provisioned to produce 400x more tokens than they could have when they were sized.
Put another way, if the singularity is coming, the first things it will destroy are those data center build outs. That doesn't matter though: the data centers are being built on circular loans, so the money needed to complete them probably never existed.
Dark times ahead; hopefully Trump will be blamed, since all this corruption happened on his watch while he had control of the senate, house and (partially) the courts. The stock-goes-down graph will show it happening after the midterms, of course.
> The stock-goes-down graph will show it happening after the midterms, of course.
This, to me, is the most interesting question in all of this. I increasingly feel like we live in a 'post-truth' world with regards to money and the stock market in particular. Like even during the COVID lockdowns the market blinked and then was at record highs within 6 months. It was just lol and completely out of touch with the reality on the ground.
The fed, and government alike, seem to have gradually shifted their mandates from trying to provide to a stable playground to trying to create a playground where important kids are never allowed to fall down. And when your existence is tied to trillions of dollars of the economy, you become the important kid.
Come to think of it this is also probably being exasperated by the rise of money management firms controlling tens of trillions of dollars. When the market valuation is technically just based on sentiment and you control that much money in a self-interested way, you're going to get weird outcomes.
The biggest customer bases will generate cash flow to build better models. The ones trailing are the ones opening their weights in order to commoditise the models. Compute providers are also incentivised to provide open models.
We will continue to have every major player trying to build vertically - model, chips, infra, packaged products and services.
> There's no lockin that Anthropic has because it's trivial for people to switch to a different provider.
This is like saying there's no lock-in that iPhone has because it's trivial to switch to Android. Or that there's no lock-in that Windows has because it's trivial to switch to Linux or Mac. Or that there's no lock-in that cars have because it's trivial to switch to bikes or trains.
They're different products. They're not the same. Being able to write the same code does not make a model Claude. Nor does being able to reach the same benchmark scores. Nor does performing one or a choice set of tasks to a comparable level of quality.
It's true, people can switch if all they want is to perform X, Y, or Z, and some other model can do X, Y and Z well enough. People switch from iPhone to Android all the time. From Windows to Linux or Mac. From cars to bikes or trains or other transit. But they're not the same thing! Android doesn't have a missing business model just because iPhone exists, nor vice versa.
You can't just say that makes cars the same as bikes or Windows the same as Linux. You can't get Claude from anybody other than Anthropic. Maybe you can get something else that does as well where it counts, but not Claude. You can't just say Anthropic is just the same as any open model provider, and that you can just switch providers for the same result. The open model has to actually deliver the desired result, because you're not going to find one that is just Claude.
This is what keeps me on Apple now. This is why most of my biggest hopes and dreams for my personal computing are that Apple will do the right things and in the right way. Because if I want to reject even one single aspect of an Apple product, it means changing to an entirely separate product. That gives up every aspect Apple had over that product at once, even the vastly positive ones. So I put up with the decisions I don't agree with, in exchange for the decisions that I do.
(Meanwhile, I plan to switch to SteamOS as soon as they support Nvidia. I don't use or care about any aspects of Windows besides running games. Now that I have Steam Frame, I don't need Windows Mixed Reality either.)
Yeah definitely, it is a thing. It varies by individual, most people have some products they won't switch out and some products they will. It all depends on what they get out of each product and what they'd get out (or not get out) of switching. Some products have characteristics I care about that nothing else replicates. Some products are slotted in and out all the time just to achieve a certain goal, or sometimes just to explore.
I don't switch from Claude for things like chatting. I gave Siri AI a big trial run when it first launched, because I absolutely love the system integration and the user interface, but the model itself is a bit dry for my needs. It doesn't really have principles the way Claude does (thanks to "constitutional AI" or whatever the fuck they call it). And I want principles, and sometimes the impression of empathy. Claude meets that bar without being obsessively sycophantic or paranoid.
(Siri AI is very good for some things, though. And I do greatly admire it. I even hacked up an adapter to use it for coding with Pi for just a short time, partly as a meme but also just out of curiosity. Just doesn't work for me here.)
I tried switching from Claude Code to Pi with DeepSeek. I had tried OpenCode at first, but it's just complete garbage. I felt so impressed of with it at first, it feels so slick on the outside, but that impression's just been shattered and shattered and shattered as I dug into it trying to fix a performance issue. Its data model and internal behavior is just not acceptable for me. (I tried Reasonix for about five minutes before concluding it's also a bit slop. It seems crafted by someone who actually understands LLMs though, unlike OpenCode, but it's still not very good.)
I absolutely loved the flexibility of Pi and the cheap prices of DeepSeek, but found myself falling into the same patterns as when trying Linux: I was constantly debugging the harness and the model, trying to get it to perform better or investigate something or other. Constantly worried that something was going wrong, that I should try to fix. And yeah, Pi has a lot of convenient features that Claude Code doesn't have, is a lot lighter, and of course can be incredibly modded. I just felt like it was also my responsibility to make my own experience, not just my capability.
At this point, Claude Code seems to just work for me, even though they removed a bunch of features I loved a while ago and never added them back (they came up with huge misunderstandings that completely missed what I actually appreciated about those features, and then proceeded to file them away as resolved). I never have to worry about debugging Claude or Claude Code.
For certain other things that aren't chatting or coding, I found that Gemini 4 Argon is actually by far the best model I've ever seen. It is the first model that's ever strictly exceeded the original November 2022 release of ChatGPT for me. Yup, it took the collective world nearly four years to replicate an experience from 2022. Gemini 3.1 Pro was so close though, I could feel it. When I first found it, I was so relieved that type of experience wasn't lost forever. Then Gemini 4 truly solidified which line of models I prefer for that use case.
Why have I been reading a variation of this comment since DS launched more than 18 months ago, and in that time ol’ No-Moat Anthropic has grown their revenue 100x, hmm?
And the cost of training is going down as a percentage of revenue over time, is it not?
They have a moat if they're like all the other businesses: unethical, buys it's moat via government protectionism, uses oligarchies to stiffle competition, etc.
https://www.wsj.com/finance/major-investors-are-skeptical-an...
I think about this because I pay $100/mo for Fable. How much more would I be willing to pay? I'm not sure yet. It depends on how smart the AI in the future becomes.
Already they have exclusive deals with trading firms who pay $$$ for faster models and more features. But there are only so many firms who will pay that much (and a lot of companies are pulling back on spend, not increasing it)
Everyone has to get capital efficient now — and fast!
It was worth $8.9 billion when the US govt got it "for free". It's worth $23B today, even after the big drops recently. How is the US public worse off?
It may only be a short-term circumstance that you need a data centre to run AI for lucrative applications. Thirty years ago you needed a mainframe to beat Kaspersky at chess. Algorithm's got better, and now you can beat the mainframe that beat Kaspersky with a bot running on a desktop Pentium of the day[1]. You could still, today, devote a full hyperscale data centre to create a chess bot to beat all other chess bots, but nobody is doing that because there's no money in it.
The frontier is continually pushed outwards. At it's outer edge you need mainframes and data-centres. Once it pushes past, you don't. There wasn't a lot of money in Chess or protein folding, despite the later having huge potential for public benefit. There is a lot of money in coding and middle management, and that's what currently has tech-bros in a frenzy. However, there's no guarantee that what the frontier moves onto next will be lucrative. It might be more chess and protein folding. If, as Algorithm's and hardware improve, coding moves to local models (Sam Altman's nightmare) and nothing lucrative immediately replaces that, what are these companies going do do with their massive data-centres?
The scale of investment we're seeing these companies commit only makes sense if they have multiple decades to reap the returns. They may well not have that much time.
_____________________
[1]https://hackaday.com/2026/11/05/pitting-a-modern-chess-engin...
Reality can’t be saturated. The set of things you can accomplish with more intelligence is unbounded. So while cheap models will continue to proliferate, there will always be a massive market at the frontier, and that won’t ever change.
Microsoft has already proven that they do not have the software chops to deliver this. Even if they had the right product managers and UI experts, MS management would torpedo the project and ensure that it gets zero adoption.
And AWS has zero software capability in that regard.
Seeing this bubble play out in real time is the most surreal thing. It's like watching a train wreck play out at a speed of a few inches per day. Like come on, just get off the track FFS.
Have you tried requesting GPU instances on Gcloud without being associated with a major institution? Yeah, good luck.
Nerfed, proprietary models have have no future and these trillion dollar valuations are laughable.
So, it's a race to the bottom because they have to keep spending more and more money on training the models just to keep ahead of the competition, while prices are driven down by the availability of open models.
Chinese companies understand this, and that's why they give their models away in the open. The real money is going to be in integrations and custom services built on top of the models. The models themselves aren't what's valuable.
This is a real question, because I just don't know. Don't the models needs lots and lots of training that need the gobs of giant data centers? So you either need the infrastructure or the money to buy computing to make the models? I see that all the money is being invested in AI and data centers but I do not quite understand why.
I am no economist or business person but AFAICT the massive amount of money being burned is because everyone in the space believes there will be a Google-like (for search) single winner who takes all when the dust settles on the AI race.
Because of this belief they are willing to burn any amount of money they can get their hands on to get to this mythical finish line as fast as they possibly can, seeing it as existential. Moving fast (particularly when your main cost is hardware capex) means moving expensively.
The real question is why anyone (let alone everyone) believes this single-winner theory in 2026 when all available evidence over the past few years shows that there is neither a moat nor a lock-in opportunity at all in this market.
For your product to be fit for purpose, it has to be able to automate your customer's ability to switch to a competing product with little to no friction. And all of the competing products are roughly the same in capabilities over a long enough timeline. Good luck building any kind of real moat with that burden.
But I guess after a certain point after you've taken on hundreds of billions of dollars of investment you're pot committed and just have to keep going through the motions even if you later see the futility, perhaps that's where the AI labs are now...? Either that or just common variety denial.
they're already making a business model where enterprises buy tokens and put them in a pool for their users to use, and I feel like that's exactly where some companies will then compare that token pricing to open source models and find it hard to justify the prices, especially as it's quite obvious that many models https://news.ycombinator.com/formatdocare now "good enough"
In the First World War, when they introduced the tank, everybody assumed its role would be to punch a big enough hole in the front lines that the horse cavalry could ride through and attack the enemy's supply lines, because That's How Wars End™. You fight the war you know, even when it's not the one you're actually fighting.
This is a really great identification of why, from an investing perspective, AI is nothing like past technologies.
That ain't going to happen, and if it somehow did in the US, things like Anna's Archive + libgen will save the rest of the planet from our self-imposed dark ages.
As for those data centers: The cost of inference and training on the same hardware is dropping much faster than Moore's Law. That multiplies with hardware advances, since Moore's Law isn't quite done yet.
So, those datacenters are depreciating at 50% every 18 months (Moore's law) at the same time as performance/$ and performance/watt of the software is dropping at some crazy rate like 10x per year. So, token per unit of data center is increasing by something like > 20x every 18 months, and that'll probably happen at least one more time. If it does, the datacenters will be provisioned to produce 400x more tokens than they could have when they were sized.
Put another way, if the singularity is coming, the first things it will destroy are those data center build outs. That doesn't matter though: the data centers are being built on circular loans, so the money needed to complete them probably never existed.
Dark times ahead; hopefully Trump will be blamed, since all this corruption happened on his watch while he had control of the senate, house and (partially) the courts. The stock-goes-down graph will show it happening after the midterms, of course.
This, to me, is the most interesting question in all of this. I increasingly feel like we live in a 'post-truth' world with regards to money and the stock market in particular. Like even during the COVID lockdowns the market blinked and then was at record highs within 6 months. It was just lol and completely out of touch with the reality on the ground.
The fed, and government alike, seem to have gradually shifted their mandates from trying to provide to a stable playground to trying to create a playground where important kids are never allowed to fall down. And when your existence is tied to trillions of dollars of the economy, you become the important kid.
Come to think of it this is also probably being exasperated by the rise of money management firms controlling tens of trillions of dollars. When the market valuation is technically just based on sentiment and you control that much money in a self-interested way, you're going to get weird outcomes.
We will continue to have every major player trying to build vertically - model, chips, infra, packaged products and services.
This is like saying there's no lock-in that iPhone has because it's trivial to switch to Android. Or that there's no lock-in that Windows has because it's trivial to switch to Linux or Mac. Or that there's no lock-in that cars have because it's trivial to switch to bikes or trains.
They're different products. They're not the same. Being able to write the same code does not make a model Claude. Nor does being able to reach the same benchmark scores. Nor does performing one or a choice set of tasks to a comparable level of quality.
It's true, people can switch if all they want is to perform X, Y, or Z, and some other model can do X, Y and Z well enough. People switch from iPhone to Android all the time. From Windows to Linux or Mac. From cars to bikes or trains or other transit. But they're not the same thing! Android doesn't have a missing business model just because iPhone exists, nor vice versa.
You can't just say that makes cars the same as bikes or Windows the same as Linux. You can't get Claude from anybody other than Anthropic. Maybe you can get something else that does as well where it counts, but not Claude. You can't just say Anthropic is just the same as any open model provider, and that you can just switch providers for the same result. The open model has to actually deliver the desired result, because you're not going to find one that is just Claude.
This is what keeps me on Apple now. This is why most of my biggest hopes and dreams for my personal computing are that Apple will do the right things and in the right way. Because if I want to reject even one single aspect of an Apple product, it means changing to an entirely separate product. That gives up every aspect Apple had over that product at once, even the vastly positive ones. So I put up with the decisions I don't agree with, in exchange for the decisions that I do.
(Meanwhile, I plan to switch to SteamOS as soon as they support Nvidia. I don't use or care about any aspects of Windows besides running games. Now that I have Steam Frame, I don't need Windows Mixed Reality either.)
I don't switch from Claude for things like chatting. I gave Siri AI a big trial run when it first launched, because I absolutely love the system integration and the user interface, but the model itself is a bit dry for my needs. It doesn't really have principles the way Claude does (thanks to "constitutional AI" or whatever the fuck they call it). And I want principles, and sometimes the impression of empathy. Claude meets that bar without being obsessively sycophantic or paranoid.
(Siri AI is very good for some things, though. And I do greatly admire it. I even hacked up an adapter to use it for coding with Pi for just a short time, partly as a meme but also just out of curiosity. Just doesn't work for me here.)
I tried switching from Claude Code to Pi with DeepSeek. I had tried OpenCode at first, but it's just complete garbage. I felt so impressed of with it at first, it feels so slick on the outside, but that impression's just been shattered and shattered and shattered as I dug into it trying to fix a performance issue. Its data model and internal behavior is just not acceptable for me. (I tried Reasonix for about five minutes before concluding it's also a bit slop. It seems crafted by someone who actually understands LLMs though, unlike OpenCode, but it's still not very good.)
I absolutely loved the flexibility of Pi and the cheap prices of DeepSeek, but found myself falling into the same patterns as when trying Linux: I was constantly debugging the harness and the model, trying to get it to perform better or investigate something or other. Constantly worried that something was going wrong, that I should try to fix. And yeah, Pi has a lot of convenient features that Claude Code doesn't have, is a lot lighter, and of course can be incredibly modded. I just felt like it was also my responsibility to make my own experience, not just my capability.
At this point, Claude Code seems to just work for me, even though they removed a bunch of features I loved a while ago and never added them back (they came up with huge misunderstandings that completely missed what I actually appreciated about those features, and then proceeded to file them away as resolved). I never have to worry about debugging Claude or Claude Code.
For certain other things that aren't chatting or coding, I found that Gemini 4 Argon is actually by far the best model I've ever seen. It is the first model that's ever strictly exceeded the original November 2022 release of ChatGPT for me. Yup, it took the collective world nearly four years to replicate an experience from 2022. Gemini 3.1 Pro was so close though, I could feel it. When I first found it, I was so relieved that type of experience wasn't lost forever. Then Gemini 4 truly solidified which line of models I prefer for that use case.
And the cost of training is going down as a percentage of revenue over time, is it not?