But the technology also doesn’t have to be transformative either. I for one am not convinced.
The two things being true may as well be: LLM will not be transformative and the LLM business is mangled in terms of ROIC vs CoC.
In fact given the behavior of AI companies, I actually consider it more likely then not that the effects of the technology is severely over-hyped. I for one do not trust the words of the people who mangle their business in terms of ROIC and CoC. Why should I?
LLM's are already transformative. I get thinking it isn't going to solve all the world problems, but, companies throughout the entire world are already using it at enormous numbers. Transformative means it changes the world, it already has. It does not mean it changes your life.
By this metric then leaded gasoline and the Deep water horizon oil spill, and Reagan era deregulations were also transformative, so was the Apollo mission, the SETI program and even Eurovision. Arguably Alchemy was also transformative even though it was unsuccessful, because the scientific funding that went into various attempts led into many vital discoveries.
If this the criteria we are using form something being transformative, then being transformative is truly unremarkable in this context, to the point of being a distraction.
If LLMs are "only" transformative in the sense of leaded gasoline or the Apollo mission, that's a massive change all by itself.
If LLMs are "transformative" in the same sense as counting "alchemy because of the scientific funding that went into various attempts led into many vital discoveries", what's going to be our version of "actually we can turn lead into gold now, we just have better things to do with the capability"?
We do have the means today to turn literal lead into literal gold.
The energy is better spent on almost anything else and the gold is radioactive afterwards, but we can do it.
The closest analogy I'd have for vibecoding is combustion engines. Historical antecedent was a toy, early industrial ones took a lot of fuel and were only useful to pump water out of the coal mines that supplied that fuel, but kept getting improved until they made a critical quality leap that took them from "slightly worse than a horse" to "marginally better than a horse" and then there were suddenly a lot of unemployed farriers.
But literal-lead-to-gold took a while longer than that, and a different set of inventions behind it.
It is becoming increasingly hard not to see it as transformative. The man hours needed to do my work has been cut down by orders of magnitude. Wether this transformation is wholly a good thing is to be seen.
The world looks pretty different today. Even if the model maker companies go out of business (I’m skeptical), the model weights would stick around (many are open freeware already).
I know many software engineers who haven’t written code this year. AI chatbots are regularly used as alternatives to searching manually by many people. Agents are becoming a valuable new enterprise tool, and now consumers tech consumers are hopping on board.
The US spends a bit less than $400B/yr creating software and the worldwide spending is around $675B/yr.
Even if you eliminate 100% of the people involved in creating software, that's still not enough money (and of course, that's not going to happen because someone has to know what to build).
Everyone is adding agents to enterprise stuff, but an overwhelming majority of the general population now hate AI for most things -- especially the "AI support" these companies are using. I think most people would rather suffer through overseas call centers with absolutely terrible representatives than deal with AI support (studies seem to indicate 80+% prefer a human to AI for support in general).
AI as a search engine is useful, but a very different animal. Proficient users want a way to check the AI like Google's AI search does (though the links sometimes don't agree with the summary), but these run very small models (8b or so) with the RAG backend doing the real work.
How many competitors does this space need? Companies can build their own proprietary RAG search engines, but users would almost always prefer the company make that public data available to Google and just use one well-optimized search engine instead of dozens of bad copies.
What about profitability? Google enshittified their search to increase retention and ad time, but AI search should reduce retention/ad time AND costs a lot more money to run too meaning it should lower their bottom line. If that weren't enough, their RAG system is almost certainly more replaceable by users with alternatives than their traditional search system. This seems like all downside for Google.
Yeah I'm not sure where I stand. It does seem like the industry keeps throwing in new hype cycles just as sentiment is about to turn on the last one. But I also can't deny some results are remarkable.
You are living in another world. I don’t know a single people who is not using ChatGPT, Gemini or Claude. Even my non tech parents are using it everyday.
How do you add them? Inside the model has tons of problems. From user context makes a bit more sense, but how do you decide and how do you ensure the context is actually beneficial for the company paying for the ads (advertisers are very sensitive about what content gets subconsciously associated with their brand).
How do you make sure a human sees them? This is a very hard problem even in normal ads and seems even more problematic with chatbots.
How do you attract advertiser dollars? Ad spending is zero-sum. You must somehow convince advertisers that your chatbot is a better ad platform than Youtube, Facebook, etc.
The only AI solution that seems reasonable is something like Google's AI search because the RAG backend ensures Google can control where ads show effectively through deterministic means (based on the websites it pulls). Google also has the ad network, consumer base, and provable humans to drive up ad value (this is without mentioning that Gemini falls behind on coding/math benchmarks, but seems to be better at "normie" interactions).
None of this helps any of the AI startups and running all this extra stuff for the same ad revenue cuts Google's profit margins too.
"Generate me an explainer video of someone holding a sign stating "I have no idea what I'm talking about" being educated about the realities of time, labor, and 'good enough'"
You're right. I didn't even need AI to get my point across.
On every objective measure, quality is improving and the trendline is clear.
I'm by no means citing that AI is perfect, or that there are no problems or challenges ahead, but the self-lobotomizing AI doomerism/rejection is turning otherwise intelligent people into a horde of anti-AI lemmings that are afraid to apply the critical thinking they believe is being lost by AI users.
These things are puppeteering blender to model things for gods sake. Are you all blind?
Compared to the Google/Facebook and other websites who run over these networks the Telcos barely capture any of the value in that exchange for themselves, whilst stumping for huge amounts of the capex
This is more like, 90% of the money from selling the bread goes to the guy slicing it, while the ones who bought, own and operate the ovens get the other 10%
If my ridiculous $100+/m subscription to the single dumb pipe that has a government-endorsed monopoly in my area is anything to go by, they're extracting plenty of value from this customer at least.
Which is northing compared to what google and other networking/application companies get from businesses to use their digital, no or little in comparison physical maintenance, products.
Also computers. DEC and IBM and Gateway and whatever didn't extract the economic benefits of digitization. Even Apple extracts a vanishing fraction of the economic benefits the iPhone has created.
Well, you can extract all the value created using your tools and inventions, because then no one would have any incentive to use them to create value.
They try to get around this by charging you more after you've invested in your tools, but even then you can only go so far (though it's astonishing how far the big guys have managed to get regardless)
One thing I'm wonder about: previous busts left useful infrastructure behind (rail, fibre) that could still be used. If/When the AI/LLM companies go bust, what will be left over afterwards and how useful will it be?
> previous busts left useful infrastructure behind (rail, fibre) that could still be used. If/When the AI/LLM companies go bust, what will be left over afterwards and how useful will it be?
I’m pretty sure that when the railroads went bust that nobody pulled up the rails and put them somewhere else. And when the internet bust happened, they didn’t pull up all the dark fiber and lay it down somewhere else.
I would bet that if all those GPUs go idle, they stay housed and cooled in the data center they were originally installed in, for better or for worse.
The numbers that Reuters describe in this article were entirely for 2025.
It's been well documented that Anthropic's revenue growth in 2026 has been enormous. This was the year of coding agents, and tokenmaxxing, and companies blowing enormous amounts of money on AI thanks to coding agent users spending hundreds (or thousands) of dollars a day.
Given that, I don't understand why the article and the headline are based exclusively on those 2025 numbers, with not so much as a hint to the reader that there are figures from the past 9 months that aren't covered by the documents Reuters saw.
Losing $8bn in 2025 isn't particularly notable if you've made ~10x that amount of revenue in 2026.
(How much did they lose in 2026? Wouldn't we love to know that!)
Is this just a thing with leaked IPO prospectuses and coverage of them?
Presumably the expenses would also be higher in 2026 as well, because inference costs scale with usage and I think training their new large models would be in there as well.
3. There is an exception to 2: if Anthropic made a gigantic discovery and was able to reduce the cost while increasing the usage - but since it would be in their best interests, they would announce this information and brag about it for months, but they haven't (what they do is just playing the quantization game all labs do, but this is beside the point).
4. Hence, we have very solid reasons to assume their costs increased.
"Opus 5.5 requires less compute to serve than Opus 5, and its pricing reflects that. Our tests show that at default settings it will cost 40% less than Opus 5 on typical workloads."
That's not a presumption, he was just being polite. It is guaranteed that inference costs increased. Unless you believe that model efficiency improvements were able to offset the entire growth of agent usage in this year, which I suppose is technically possible although it would be a truly remarkable achievement.
Reuters probably saving 2026 for another writeup. To be honest, this is nothing surprising for 2025. I think we all expected unsurmountable costs. AI has always been and still is a huge bet into the future. The future is not written.
>It's been well documented that Anthropic's revenue growth in 2026 has been enormous.
Has it? Don't they just self-leak/press release 'ARR' revenue claims? Has there been any audited and formal filings? Including all the off-balance sheet and other creative stuff
We've also heard plenty of stories from companies like Uber about a need to set a limit on their token spending because they were spending so much on Anthropic. That's a solid sign that the revenue numbers are real.
What Uber and other companies said is that the money they spent didn't generate the expected value they were promised. If it had generated $2 for every $1 spent, then Uber would be spending with no limits.
The lack of demonstrable return on investment means there is a ceiling to the market size, and profit is capped as cheaper models and self hosted open source models limit the amount flagship models can charge.
Long way to go till market saturation though so they will grow for foreseeable, but its a quandary of how much they spend on R&D vs giving future shareholders dividends.
I didn't mean 'no ROI', clearly AI is a productivity boost.
For the AI companies to IPO at the money they want, they have to promise that they are going to be worth many multiples more of that value.
But currently they are spending hundreds of billions and despite this their models are not light years ahead of each other, or of the companies with much lower budgets and compute power.
What they claim they can offer in future profits is subject to considerable debate.
It's in the interests of VC's and their underlying rich investors to hype it ip, IPO and bank the cold hard cash.
If Uber caps employee AI spend at $1500, then they've just admitted per employee AI spend on par with premium health care or commercial rent both in the 1000-1500 range, and this limited to just technical employees, i.e. subsect so presumably AI as % of business cost even lower. Health insurance 1T industry... commercial RE like 100 billion...
Anthropic thinks it's worth more than insuring all US workers AND putting roof over their heads. AI sector thinks its worth magnitude more in aggregate. If they fully displace people, i.e. untied to headcount eventually then sure, but if not, based on what Uber pays, AI is basically worth as much as insurance.
Simon you're really going to need to float the numbers for us if you're going to make claims about enormous revenue. And don't forget to include things like cost of revenue as well !
I assembled some of those numbers in May, when they claimed they had grown from $9bn in annualized revenue in December 2025 to $47bn in early May. https://simonwillison.net/2026/May/29/anthropic/
And sure, they might be lying about those figures - but if they are, that's investor fraud, and they'll be in hot water with the SEC when they try to IPO. I don't think they are lying about the figures.
If I had numbers on their cost of revenue I would share those. As it stands I'm going to have to wait for either more leaks or their S-1.
This isn't a standardized metric like GAAP revenue, and the company chooses how to compute it. Nothing stops them from taking a strong month (or week, or day, or hell even a second of revenue) and multiplying it out. It's also a different thing from booked revenue: the prospectus shows about $4.6B recognized in 2025, versus a ~$9B run rate at year-end.
Even taken at face value, it tells us nothing about costs. In 2025 they lost about $8B operating on $4.6B revenue, with compute alone at $7.3B. The $42B net loss is inflated by a ~$34B non-cash charge, so I'm ignoring that. The company says it hit positive adjusted operating income in Q2, but that's their own number, "adjusted" excludes whatever they choose such as training costs or the cost of staff, and we won't know the real picture until the full S-1 is public. Meanwhile they've disclosed $518B in future compute and infrastructure commitments against ~$20B in cash.
If they had a trillion dollars worth of revenue, it wouldn't mean jack shit if they had a trillion + 1 in losses.
> If they had a trillion dollars worth of revenue, it wouldn't mean jack shit if they had a trillion + 1 in losses.
I don't understand that argument.
If a company has a trillion dollars in revenue, even if they are losing money hand-over-fist, that still means they have convinced other companies to cough up a trillion dollars for what they are selling. That's a big deal!
The only case that isn't impressive is if they are literally selling dollar bills for 90 cents.
You can argue that Anthropic are subsidizing their tokens all you want, but since as a customer you can't just turn around and sell a token yourself for more than you paid for it that's still not a good argument for dismissing the amount people are willing to spend.
That was my question too. 2025 is really old in AI-company time. But I'm suspecting the "leak" was on purpose so they can now come and show how much better (but still not great) the first half of 2026 was.
You know the losses increased even more massively or Anthropic itself would have leaked the number already...
Most interesting: "Anthropic said nearly a quarter of its revenue came from two customers last year, and as part of its risk factors, warned that many of its largest clients were not locked into long-term contracts and could cut or stop spending."
But that was in 2025. It's not surprising to hear that in 2025 they had two customers responsible for a quarter of their revenue (those companies were Cursor and GitHub Copilot btw) - they hadn't yet sold Claude Code packages to vast numbers of large companies.
Here's another amazing trick for revenue growth: buy items on Amazon and then sell them on your website for half the price.
Revenue growth means nothing if you don't have a viable business plan. Anthropic doesn't have one, except "this thing is more important than Big Bang". I don't discount the possibility that they stay afloat on borrowed money long enough to actually find a way to be profitable / rule the world. But it's nothing other than appeal to consequence right now.
So either your analysis is correct or the scientists begging and pleading with people to finally acknowledge what they've been saying since 2023 are correct. I know where I'll be putting my money!
I don't understand, scientists were pleading that Anthropic will be profitable? Doomsday predictions around AI have nothing to do with the financial outlook of a specific company.
I was just replying to your framing, sorry to be unclear. You said that Anthropic is counting on AI transforming the whole economy/be the "big bang". I'm saying that that bet is extremely rational -- to bet otherwise is to ignore science because its conclusions make you anxious.
does anyone understand what this part actually means?
"The near-$42 billion net loss included a roughly $34 billion accounting charge that reflected an increase in the estimated value of financing that could eventually turn into Anthropic shares, rather than money the company spent running its business."
If you truly believe that the cost of financing should not be included in your net loss (I don‘t know anything about money, so I don‘t know whether you should or not) $8 billion in net loss is also mindboggingly large number.
As a member of the working class who believes this money is coming from systemic exploitation of the working classes, I hate it that Anthrhopic can just loose $8 billion like that. Those $8 billion could have been used elsewhere, including to mitigate the effects of the climate disaster, to educate the youth, towards social security, or to increase the wages of the working people who could have used it to pay rent, or go to the cinema, or a family trip to visit their grandma who lives in Wyoming.
Okay. About 50 million households in America earning under $50k.
8 billion is enough to give each of them $160. Great, all of Anthropic’s spending is enough to give each family a trip to Texas Roadhouse including dessert.
34.5M Americans below the poverty line. Divide out that $8B and it's a little over $230 per person.
Have you ever been in poverty? I grew up that way and the then-equivalent of $230 would have meant not skipping Christmas when I was 6. It would have meant a full refrigerator, gas for the car for a couple of months, or so many other things.
they don't light it on fire. it circulates in the economy. plenty of working class people doing enormously well from the data centre, chip production build out.
Money would circulate at least as much if used for climate solutions and even more so if given to workers. The AI build out is not benefitting workers, like hardly at all, it creates a dismal amount of jobs, and is raising utility prices for many of us.
If I understand correctly it means that they have a lot of debt and with increasing interest rates, they have to spend a significant chunk in just paying interest
Not $518B in the next year, they plan "to spend $518 billion on cloud, computing and infrastructure obligations in coming years" (number of years unspecified)
Looking at Nvidia sales and even the most optimistic "use this GPU for 6 years despite its horrible inefficiency", you still get a very big number per-year.
Google, Amazon, Oracle, Microsoft, Meta, SpaceX, and tons of other startups are spending trillions on data centers, but everyone is renting them out. They aren't renting them to each other (and paying the extra overhead when they have their own servers).
That leaves basically just OpenAI and Anthropic on the hook to pay for everything BEFORE the GPU half of it depreciates away, but they are busy cutting prices to compete with Chinese models which runs counter to their need to increase prices to fulfill their obligations.
How would it help them? The real competition is coming from China and they don't need financials to know they are stomping Anthropic in all these areas.
Keeping financiers of their IPO from understanding what they might be buying seems to swerve into fraud territory. Keeping their investors from knowing/understanding what the problems are also seems fraudulent.
> How would it help them? The real competition is coming from China
I mean, sure, you think that and I know that many others do too.
It's Anthropic and OpenAI mainly competing for who gets used by the US government and possibly any US business that does any contracting work, while many others get banned. So basically, Fortune 500+.
There is also a non-zero possibility that the US Gov nationalizes the entire industry and then starts directly throwing Trillions USD into it in the name of "national security". Low % chance, but I can see it happening. Or at least the Fed Gov taking a controlling ownership stake.
There won’t be a bailout. I doubt the obligations will even amount to anything substantial anyways.
Many of these obligations are with the big clouds who don’t have capacity to serve them anyways. Google has said they have hundreds of billions in purchase obligations they can’t fill because they don’t have enough TPUs and data centers. Google gets a bigger marketing and investor headline, and there’s a queue behind Anthropic if they bail, so of course everyone would sign the commitments knowing the risks.
I’m sure AWS and Azure are on the same page.
In the worst case, many of their biggest customers have tons of GPUs (Meta, AWS), so they could always license out their raw models at steep discount to help absolve themselves of the obligations.
Finally, in the worst case scenario, even with margin compression, they pre-purchased most of the compute so there’s not enough for the open models to run on. If Opus was the same price as Kimi/GLM, I doubt many people would pick the open models.
Yeah not sure about that. OpenAI and oracle are already losing money because of their obligations. They are already going to the worst case scenario and it's worse than you describe.
> Anthropic said nearly a quarter of its revenue came from two customers last year, and as part of its risk factors, warned that many of its largest clients were not locked into long-term contracts and could cut or stop spending.
Is it possible those are cloud providers? Someone using Claude on google vertex might just switch to Claude on AWS or Anthropic. Also I don’t imagine cloud providers suddenly cutting the cord.
Why xAI? They have no market share and their own data centers are over 90% idle/rented out to the other AI shops, so it's more likely that the money flows in the opposite direction.
Gaap net loss almost 42B on less than 5B of revenue, I cant even imagine. Where does all the money go? If the name attached to this was anything other than Anthropic these would be some seriously horrible financials, and that's before getting into all the colorful accounting they're using.
> the gross margins right now are very positive. [...] So the company loses money. Each model makes money, but the company loses money.
Amodei mentioned the reason for the big losses in an interview released on Feb 2026[1]. Assuming he was telling the truth, that's not a bad position to be in. From what I can tell from the replies to simonw, a lot of people are cynical about Anthropic's growth potential. I'd like to encourage the cynics to bet against Anthropic wish them luck. I will be doing the opposite.
This is the part that really surprised me: "Anthropic said nearly a quarter of its revenue came from two customers last year".
Who are these two customers? A quick Google search seems to indicate that Cognizant is their largest "enterprise" customer by number of seats, Microsoft is their largest API customer, and the $10B contract they recently signed with Meta is their largest contract ever. But those metrics don't necessarily translate directly into annual revenue.
2 things can be true:
-LLMs profoundly change society -The LLM business is mangled in terms of ROIC vs CoC
Such a business already exists - airlines.
This nuance is what many here refuse/find difficult to understand.
Well said, the technology can be transformative while being structurally challenging to build a viable business around.
But the technology also doesn’t have to be transformative either. I for one am not convinced.
The two things being true may as well be: LLM will not be transformative and the LLM business is mangled in terms of ROIC vs CoC.
In fact given the behavior of AI companies, I actually consider it more likely then not that the effects of the technology is severely over-hyped. I for one do not trust the words of the people who mangle their business in terms of ROIC and CoC. Why should I?
LLM's are already transformative. I get thinking it isn't going to solve all the world problems, but, companies throughout the entire world are already using it at enormous numbers. Transformative means it changes the world, it already has. It does not mean it changes your life.
By this metric then leaded gasoline and the Deep water horizon oil spill, and Reagan era deregulations were also transformative, so was the Apollo mission, the SETI program and even Eurovision. Arguably Alchemy was also transformative even though it was unsuccessful, because the scientific funding that went into various attempts led into many vital discoveries.
If this the criteria we are using form something being transformative, then being transformative is truly unremarkable in this context, to the point of being a distraction.
I just will never understand how you all can have your head so deep in the sand.
Attempting to understand perspectives that feel alien to you will be good for your personal growth. (Try asking your LLM.)
If LLMs are "only" transformative in the sense of leaded gasoline or the Apollo mission, that's a massive change all by itself.
If LLMs are "transformative" in the same sense as counting "alchemy because of the scientific funding that went into various attempts led into many vital discoveries", what's going to be our version of "actually we can turn lead into gold now, we just have better things to do with the capability"?
https://en.wikipedia.org/wiki/Nuclear_transmutation
"actually we can turn lead into gold now"
Isn't it vibecoding? Turn out it's not really gold
No, I don't think so.
We do have the means today to turn literal lead into literal gold.
The energy is better spent on almost anything else and the gold is radioactive afterwards, but we can do it.
The closest analogy I'd have for vibecoding is combustion engines. Historical antecedent was a toy, early industrial ones took a lot of fuel and were only useful to pump water out of the coal mines that supplied that fuel, but kept getting improved until they made a critical quality leap that took them from "slightly worse than a horse" to "marginally better than a horse" and then there were suddenly a lot of unemployed farriers.
But literal-lead-to-gold took a while longer than that, and a different set of inventions behind it.
Don’t. It doesn’t matter. Join Ed Zitron and shout at clouds while the world passes you by.
Would love to, but… https://bsky.app/profile/smoothdunk2.bsky.social/post/3mvx47...
It is becoming increasingly hard not to see it as transformative. The man hours needed to do my work has been cut down by orders of magnitude. Wether this transformation is wholly a good thing is to be seen.
Do you not feel we’re already transformed?
The world looks pretty different today. Even if the model maker companies go out of business (I’m skeptical), the model weights would stick around (many are open freeware already).
I know many software engineers who haven’t written code this year. AI chatbots are regularly used as alternatives to searching manually by many people. Agents are becoming a valuable new enterprise tool, and now consumers tech consumers are hopping on board.
The US spends a bit less than $400B/yr creating software and the worldwide spending is around $675B/yr.
Even if you eliminate 100% of the people involved in creating software, that's still not enough money (and of course, that's not going to happen because someone has to know what to build).
Everyone is adding agents to enterprise stuff, but an overwhelming majority of the general population now hate AI for most things -- especially the "AI support" these companies are using. I think most people would rather suffer through overseas call centers with absolutely terrible representatives than deal with AI support (studies seem to indicate 80+% prefer a human to AI for support in general).
AI as a search engine is useful, but a very different animal. Proficient users want a way to check the AI like Google's AI search does (though the links sometimes don't agree with the summary), but these run very small models (8b or so) with the RAG backend doing the real work.
How many competitors does this space need? Companies can build their own proprietary RAG search engines, but users would almost always prefer the company make that public data available to Google and just use one well-optimized search engine instead of dozens of bad copies.
What about profitability? Google enshittified their search to increase retention and ad time, but AI search should reduce retention/ad time AND costs a lot more money to run too meaning it should lower their bottom line. If that weren't enough, their RAG system is almost certainly more replaceable by users with alternatives than their traditional search system. This seems like all downside for Google.
Yeah I'm not sure where I stand. It does seem like the industry keeps throwing in new hype cycles just as sentiment is about to turn on the last one. But I also can't deny some results are remarkable.
Frankly, outside of coding there's no much real use for AI, which is basically NLP.
There's, uh, a reason ChatGPT is one of the biggest consumer apps of all time.
That nobody uses on their phones. It’s like that install of angrybirds everyone has that they haven’t touched in a decade
You are living in another world. I don’t know a single people who is not using ChatGPT, Gemini or Claude. Even my non tech parents are using it everyday.
As of today, ChatGPT is the #4 Top Free app on the US iOS App Store: this chart only measures new/recent downloads.
The #1 and #2 apps are non-coding, pure consumer AI applications (Meta Muse and Momo).
How many people are going to pay a 20 dollars subscription for it?
That is not relevant to your assertion of "AI has no real use besides NLP."
apparently very few people - I know tons of people who use it and almost nobody who pays for it, particularly outside of tech
Well that's why they're going to start injecting ads.
Ads in AI are problematic in a big way.
How do you add them? Inside the model has tons of problems. From user context makes a bit more sense, but how do you decide and how do you ensure the context is actually beneficial for the company paying for the ads (advertisers are very sensitive about what content gets subconsciously associated with their brand).
How do you make sure a human sees them? This is a very hard problem even in normal ads and seems even more problematic with chatbots.
How do you attract advertiser dollars? Ad spending is zero-sum. You must somehow convince advertisers that your chatbot is a better ad platform than Youtube, Facebook, etc.
The only AI solution that seems reasonable is something like Google's AI search because the RAG backend ensures Google can control where ads show effectively through deterministic means (based on the websites it pulls). Google also has the ad network, consumer base, and provable humans to drive up ad value (this is without mentioning that Gemini falls behind on coding/math benchmarks, but seems to be better at "normie" interactions).
None of this helps any of the AI startups and running all this extra stuff for the same ad revenue cuts Google's profit margins too.
"Generate me an explainer video of someone holding a sign stating "I have no idea what I'm talking about" being educated about the realities of time, labor, and 'good enough'"
You're right. I didn't even need AI to get my point across.
If that's your idea of "real use", you're really reinforcing their point.
Explainer videos aren't real use?
I mean, if self education isn't a "real use", I wonder why all those libraries decided to dedicate all that shelf space to nonfiction books?
Wikipedia could probably save a lot of hard drive space too.
Absolute slopshit "explainers" are not real use, no.
On every objective measure, quality is improving and the trendline is clear.
I'm by no means citing that AI is perfect, or that there are no problems or challenges ahead, but the self-lobotomizing AI doomerism/rejection is turning otherwise intelligent people into a horde of anti-AI lemmings that are afraid to apply the critical thinking they believe is being lost by AI users.
These things are puppeteering blender to model things for gods sake. Are you all blind?
> On every objective measure, quality is improving and the trendline is clear.
Where? I can't find one company that did "AI layoffs" and their app/services got better. Give me just one good example please
Funny, I don’t remember NLP being useful as a coding agent or solving cutting edge math problems.
Isn't telecom kinda the same? Lots and lots of capex, turned into dumb pipes that can't extract value from their customers.
I don't think telecom is the same. 2025 net income for the big 3 US telecoms: AT&T: $22B Verizon: $17B T-Mobile: $11B
Compared to the Google/Facebook and other websites who run over these networks the Telcos barely capture any of the value in that exchange for themselves, whilst stumping for huge amounts of the capex
Meta alone made more than those three combined at $60B, Google made $132B
Well, it would be weird if a loaf of bread cost less then the wheat it's made from.
This is more like, 90% of the money from selling the bread goes to the guy slicing it, while the ones who bought, own and operate the ovens get the other 10%
If my ridiculous $100+/m subscription to the single dumb pipe that has a government-endorsed monopoly in my area is anything to go by, they're extracting plenty of value from this customer at least.
Which is northing compared to what google and other networking/application companies get from businesses to use their digital, no or little in comparison physical maintenance, products.
> Such a business already exists - airlines
Also computers. DEC and IBM and Gateway and whatever didn't extract the economic benefits of digitization. Even Apple extracts a vanishing fraction of the economic benefits the iPhone has created.
Well, you can extract all the value created using your tools and inventions, because then no one would have any incentive to use them to create value.
They try to get around this by charging you more after you've invested in your tools, but even then you can only go so far (though it's astonishing how far the big guys have managed to get regardless)
what everyone is missing from the example: neither airline nor telecom requires a trillion dollar investments to get the same revenue
> -LLMs profoundly change society -The LLM business is mangled in terms of ROIC vs CoC
People seem to have forgotten the bust in railroads and Internet (etc), even though those technologies were transformative:
* https://en.wikipedia.org/wiki/Technological_Revolutions_and_...
One thing I'm wonder about: previous busts left useful infrastructure behind (rail, fibre) that could still be used. If/When the AI/LLM companies go bust, what will be left over afterwards and how useful will it be?
Seems like it might leave behind underutilized data centers full of subsidized compute seeking paying tenants.
> previous busts left useful infrastructure behind (rail, fibre) that could still be used. If/When the AI/LLM companies go bust, what will be left over afterwards and how useful will it be?
Many many idle GPUs.
Hey, you can do a lot of fun and useful things with idle GPUs if the prices come down! Think of all the RAM as well.
Who will house and cool them?
I’m pretty sure that when the railroads went bust that nobody pulled up the rails and put them somewhere else. And when the internet bust happened, they didn’t pull up all the dark fiber and lay it down somewhere else.
I would bet that if all those GPUs go idle, they stay housed and cooled in the data center they were originally installed in, for better or for worse.
I'm confused.
The numbers that Reuters describe in this article were entirely for 2025.
It's been well documented that Anthropic's revenue growth in 2026 has been enormous. This was the year of coding agents, and tokenmaxxing, and companies blowing enormous amounts of money on AI thanks to coding agent users spending hundreds (or thousands) of dollars a day.
Given that, I don't understand why the article and the headline are based exclusively on those 2025 numbers, with not so much as a hint to the reader that there are figures from the past 9 months that aren't covered by the documents Reuters saw.
Losing $8bn in 2025 isn't particularly notable if you've made ~10x that amount of revenue in 2026.
(How much did they lose in 2026? Wouldn't we love to know that!)
Is this just a thing with leaked IPO prospectuses and coverage of them?
Presumably the expenses would also be higher in 2026 as well, because inference costs scale with usage and I think training their new large models would be in there as well.
I’m not sure “presumption” is a valid methodology to evaluate a company’s performance.
What methodology would you propose given the available information?
Gather more information before coming to a conclusion. Otherwise, do not jump to a conclusion prematurely.
Let me offer an alternative: logical reasoning.
1. Increased revenue normally means increased usage. [We don't dispute that, right?]
2. Increased usage normally means increased cost.
3. There is an exception to 2: if Anthropic made a gigantic discovery and was able to reduce the cost while increasing the usage - but since it would be in their best interests, they would announce this information and brag about it for months, but they haven't (what they do is just playing the quantization game all labs do, but this is beside the point).
4. Hence, we have very solid reasons to assume their costs increased.
https://www.anthropic.com/claude-opus-5-5
"Opus 5.5 requires less compute to serve than Opus 5, and its pricing reflects that. Our tests show that at default settings it will cost 40% less than Opus 5 on typical workloads."
https://www.anthropic.com/claude-sonnet-5-5
"Sonnet 5.5 requires fewer tokens per task than Sonnet 5, so it’s less expensive to run. It also generates output 30%+ faster"
OpenAI have been achieving even more impressive optimizations, hence why GPT-6 Sol and GPT-6 Luna are half the price of their 5.6 equivalents.
That's not a presumption, he was just being polite. It is guaranteed that inference costs increased. Unless you believe that model efficiency improvements were able to offset the entire growth of agent usage in this year, which I suppose is technically possible although it would be a truly remarkable achievement.
We know inference costs scale with usage. Unless they found a way to make it much more efficient (which is possible). I think it's a fair presumption.
Not including Q1 and Q2 numbers sounds like a hit piece.
Reuters probably saving 2026 for another writeup. To be honest, this is nothing surprising for 2025. I think we all expected unsurmountable costs. AI has always been and still is a huge bet into the future. The future is not written.
Has it? Don't they just self-leak/press release 'ARR' revenue claims? Has there been any audited and formal filings? Including all the off-balance sheet and other creative stuff
Trust me bro ... enormous ... and we are going to destroy the human race too.
...definitely!
We've also heard plenty of stories from companies like Uber about a need to set a limit on their token spending because they were spending so much on Anthropic. That's a solid sign that the revenue numbers are real.
It's a solid sign the numbers are not real...
What Uber and other companies said is that the money they spent didn't generate the expected value they were promised. If it had generated $2 for every $1 spent, then Uber would be spending with no limits.
The lack of demonstrable return on investment means there is a ceiling to the market size, and profit is capped as cheaper models and self hosted open source models limit the amount flagship models can charge.
Long way to go till market saturation though so they will grow for foreseeable, but its a quandary of how much they spend on R&D vs giving future shareholders dividends.
If Uber think there's no ROI, why do they allow each of their employees to spend $1,500 per month per tool?
Shouldn't they have set that per-employee budget to zero instead?
(I dug up the original source for that Uber doubts the ROI story a few months ago, it's a lot weaker than the headlines about it suggested: https://simonwillison.net/2026/May/27/product-market-fit/#th...)
I didn't mean 'no ROI', clearly AI is a productivity boost.
For the AI companies to IPO at the money they want, they have to promise that they are going to be worth many multiples more of that value.
But currently they are spending hundreds of billions and despite this their models are not light years ahead of each other, or of the companies with much lower budgets and compute power.
What they claim they can offer in future profits is subject to considerable debate.
It's in the interests of VC's and their underlying rich investors to hype it ip, IPO and bank the cold hard cash.
Yeah, I agree with all of that.
If Uber caps employee AI spend at $1500, then they've just admitted per employee AI spend on par with premium health care or commercial rent both in the 1000-1500 range, and this limited to just technical employees, i.e. subsect so presumably AI as % of business cost even lower. Health insurance 1T industry... commercial RE like 100 billion...
Anthropic thinks it's worth more than insuring all US workers AND putting roof over their heads. AI sector thinks its worth magnitude more in aggregate. If they fully displace people, i.e. untied to headcount eventually then sure, but if not, based on what Uber pays, AI is basically worth as much as insurance.
Simon you're really going to need to float the numbers for us if you're going to make claims about enormous revenue. And don't forget to include things like cost of revenue as well !
I assembled some of those numbers in May, when they claimed they had grown from $9bn in annualized revenue in December 2025 to $47bn in early May. https://simonwillison.net/2026/May/29/anthropic/
Since then they've reported $65bn in annualized revenue by July: https://simonwillison.net/2026/Aug/23/anthropics-best-ai-mod...
And sure, they might be lying about those figures - but if they are, that's investor fraud, and they'll be in hot water with the SEC when they try to IPO. I don't think they are lying about the figures.
If I had numbers on their cost of revenue I would share those. As it stands I'm going to have to wait for either more leaks or their S-1.
> It's been well documented that Anthropic's revenue growth in 2026 has been enormous.
By whom, exactly?
Every half-respectable programmer has at least the pro subscription.
What for, exactly?
Primarily forgetting how to be a half-respectable programmer.
https://www.cnbc.com/2026/08/17/anthropic-says-annualized-re...
https://www.reuters.com/technology/anthropic-revenue-run-rat...
> annualized revenue run rate
This isn't a standardized metric like GAAP revenue, and the company chooses how to compute it. Nothing stops them from taking a strong month (or week, or day, or hell even a second of revenue) and multiplying it out. It's also a different thing from booked revenue: the prospectus shows about $4.6B recognized in 2025, versus a ~$9B run rate at year-end.
Even taken at face value, it tells us nothing about costs. In 2025 they lost about $8B operating on $4.6B revenue, with compute alone at $7.3B. The $42B net loss is inflated by a ~$34B non-cash charge, so I'm ignoring that. The company says it hit positive adjusted operating income in Q2, but that's their own number, "adjusted" excludes whatever they choose such as training costs or the cost of staff, and we won't know the real picture until the full S-1 is public. Meanwhile they've disclosed $518B in future compute and infrastructure commitments against ~$20B in cash.
If they had a trillion dollars worth of revenue, it wouldn't mean jack shit if they had a trillion + 1 in losses.
> If they had a trillion dollars worth of revenue, it wouldn't mean jack shit if they had a trillion + 1 in losses.
I don't understand that argument.
If a company has a trillion dollars in revenue, even if they are losing money hand-over-fist, that still means they have convinced other companies to cough up a trillion dollars for what they are selling. That's a big deal!
The only case that isn't impressive is if they are literally selling dollar bills for 90 cents.
You can argue that Anthropic are subsidizing their tokens all you want, but since as a customer you can't just turn around and sell a token yourself for more than you paid for it that's still not a good argument for dismissing the amount people are willing to spend.
That was my question too. 2025 is really old in AI-company time. But I'm suspecting the "leak" was on purpose so they can now come and show how much better (but still not great) the first half of 2026 was.
You know the losses increased even more massively or Anthropic itself would have leaked the number already...
Most interesting: "Anthropic said nearly a quarter of its revenue came from two customers last year, and as part of its risk factors, warned that many of its largest clients were not locked into long-term contracts and could cut or stop spending."
But that was in 2025. It's not surprising to hear that in 2025 they had two customers responsible for a quarter of their revenue (those companies were Cursor and GitHub Copilot btw) - they hadn't yet sold Claude Code packages to vast numbers of large companies.
Some of these numbers are crazy. The scale of them is hard to comprehend.
- $42 billion net loss in 2025
- $518 billion in infrastructure obligations coming up (EDIT)
- 1/4 of revenue coming from 2 customers
Sure, but their revenue growth numbers are also crazy.
Here's another amazing trick for revenue growth: buy items on Amazon and then sell them on your website for half the price.
Revenue growth means nothing if you don't have a viable business plan. Anthropic doesn't have one, except "this thing is more important than Big Bang". I don't discount the possibility that they stay afloat on borrowed money long enough to actually find a way to be profitable / rule the world. But it's nothing other than appeal to consequence right now.
So either your analysis is correct or the scientists begging and pleading with people to finally acknowledge what they've been saying since 2023 are correct. I know where I'll be putting my money!
I don't understand, scientists were pleading that Anthropic will be profitable? Doomsday predictions around AI have nothing to do with the financial outlook of a specific company.
I was just replying to your framing, sorry to be unclear. You said that Anthropic is counting on AI transforming the whole economy/be the "big bang". I'm saying that that bet is extremely rational -- to bet otherwise is to ignore science because its conclusions make you anxious.
Anthropic is betting they will capture this market, not that it exists.
does anyone understand what this part actually means?
"The near-$42 billion net loss included a roughly $34 billion accounting charge that reflected an increase in the estimated value of financing that could eventually turn into Anthropic shares, rather than money the company spent running its business."
They increased the projected liability of pre-existing financial deals that include share options.
If you truly believe that the cost of financing should not be included in your net loss (I don‘t know anything about money, so I don‘t know whether you should or not) $8 billion in net loss is also mindboggingly large number.
As a member of the working class who believes this money is coming from systemic exploitation of the working classes, I hate it that Anthrhopic can just loose $8 billion like that. Those $8 billion could have been used elsewhere, including to mitigate the effects of the climate disaster, to educate the youth, towards social security, or to increase the wages of the working people who could have used it to pay rent, or go to the cinema, or a family trip to visit their grandma who lives in Wyoming.
Okay. About 50 million households in America earning under $50k.
8 billion is enough to give each of them $160. Great, all of Anthropic’s spending is enough to give each family a trip to Texas Roadhouse including dessert.
34.5M Americans below the poverty line. Divide out that $8B and it's a little over $230 per person.
Have you ever been in poverty? I grew up that way and the then-equivalent of $230 would have meant not skipping Christmas when I was 6. It would have meant a full refrigerator, gas for the car for a couple of months, or so many other things.
they don't light it on fire. it circulates in the economy. plenty of working class people doing enormously well from the data centre, chip production build out.
Money would circulate at least as much if used for climate solutions and even more so if given to workers. The AI build out is not benefitting workers, like hardly at all, it creates a dismal amount of jobs, and is raising utility prices for many of us.
This is likely a convertible debt investment from Amazon being revalued. The numbers are pretty close, there might be some smaller notes in there too.
https://www.sec.gov/Archives/edgar/data/1018724/000101872426...
Convertible notes are worth more as the valuation of the company increases and under accounting rules that’s considered an expense for the company?
Because it is, because every other shareholder gets diluted.
If I understand correctly it means that they have a lot of debt and with increasing interest rates, they have to spend a significant chunk in just paying interest
I guess those 2 customers must be state actors as one of them must be spending at least 12B$ ?
Not $518B in the next year, they plan "to spend $518 billion on cloud, computing and infrastructure obligations in coming years" (number of years unspecified)
Looking at Nvidia sales and even the most optimistic "use this GPU for 6 years despite its horrible inefficiency", you still get a very big number per-year.
Google, Amazon, Oracle, Microsoft, Meta, SpaceX, and tons of other startups are spending trillions on data centers, but everyone is renting them out. They aren't renting them to each other (and paying the extra overhead when they have their own servers).
That leaves basically just OpenAI and Anthropic on the hook to pay for everything BEFORE the GPU half of it depreciates away, but they are busy cutting prices to compete with Chinese models which runs counter to their need to increase prices to fulfill their obligations.
I feel like we can't really evaluate them until we see a fairly detailed breakdown that shows inference margin, training costs and non-compute R&D
This is true. If they aren't showing the details, it's because the details make them look worse.
Not necessarily, it could help potential competitors or financers to better understand the market / competition.
That's true, my comment was not correct.
How would it help them? The real competition is coming from China and they don't need financials to know they are stomping Anthropic in all these areas.
Keeping financiers of their IPO from understanding what they might be buying seems to swerve into fraud territory. Keeping their investors from knowing/understanding what the problems are also seems fraudulent.
> How would it help them? The real competition is coming from China
I mean, sure, you think that and I know that many others do too.
It's Anthropic and OpenAI mainly competing for who gets used by the US government and possibly any US business that does any contracting work, while many others get banned. So basically, Fortune 500+.
There is also a non-zero possibility that the US Gov nationalizes the entire industry and then starts directly throwing Trillions USD into it in the name of "national security". Low % chance, but I can see it happening. Or at least the Fed Gov taking a controlling ownership stake.
It's also possible that the US Federal Government forces NVidia to outright prevent open models via: https://venturebeat.com/infrastructure/nvidias-openshell-con...
Factor in their margin being chewed through with every open model release.
That spending obligation sounds extremely unlikely to be fulfilled without a bailout. Anthropic's IPO is them passing the hot potato forward.
There won’t be a bailout. I doubt the obligations will even amount to anything substantial anyways.
Many of these obligations are with the big clouds who don’t have capacity to serve them anyways. Google has said they have hundreds of billions in purchase obligations they can’t fill because they don’t have enough TPUs and data centers. Google gets a bigger marketing and investor headline, and there’s a queue behind Anthropic if they bail, so of course everyone would sign the commitments knowing the risks.
I’m sure AWS and Azure are on the same page.
In the worst case, many of their biggest customers have tons of GPUs (Meta, AWS), so they could always license out their raw models at steep discount to help absolve themselves of the obligations.
Finally, in the worst case scenario, even with margin compression, they pre-purchased most of the compute so there’s not enough for the open models to run on. If Opus was the same price as Kimi/GLM, I doubt many people would pick the open models.
Yeah not sure about that. OpenAI and oracle are already losing money because of their obligations. They are already going to the worst case scenario and it's worse than you describe.
> Anthropic said nearly a quarter of its revenue came from two customers last year, and as part of its risk factors, warned that many of its largest clients were not locked into long-term contracts and could cut or stop spending.
Is it possible those are cloud providers? Someone using Claude on google vertex might just switch to Claude on AWS or Anthropic. Also I don’t imagine cloud providers suddenly cutting the cord.
I would assume clouds count as resellers not customers.
One of those customers is the US government, I wonder who the other one is
Likely Meta
Nah, more probably some VC backed coding tool startup.
Probably cursor.
No.
I'm guessing XAI
Why xAI? They have no market share and their own data centers are over 90% idle/rented out to the other AI shops, so it's more likely that the money flows in the opposite direction.
Amazon is my guess
Gaap net loss almost 42B on less than 5B of revenue, I cant even imagine. Where does all the money go? If the name attached to this was anything other than Anthropic these would be some seriously horrible financials, and that's before getting into all the colorful accounting they're using.
> the gross margins right now are very positive. [...] So the company loses money. Each model makes money, but the company loses money.
Amodei mentioned the reason for the big losses in an interview released on Feb 2026[1]. Assuming he was telling the truth, that's not a bad position to be in. From what I can tell from the replies to simonw, a lot of people are cynical about Anthropic's growth potential. I'd like to encourage the cynics to bet against Anthropic wish them luck. I will be doing the opposite.
[1] https://www.dwarkesh.com/p/dario-amodei-2
Why are they only revealing 2025 numbers 3 quarters into 2026?
I assume routers is going to drop the H1 2026 numbers later to keep the hype going
This is the part that really surprised me: "Anthropic said nearly a quarter of its revenue came from two customers last year".
Who are these two customers? A quick Google search seems to indicate that Cognizant is their largest "enterprise" customer by number of seats, Microsoft is their largest API customer, and the $10B contract they recently signed with Meta is their largest contract ever. But those metrics don't necessarily translate directly into annual revenue.
Reuters source who leaked them the info only showed them the 2025 numbers.
https://www.reuters.com/business/retail-consumer/anthropic-t...
FYI. Not sure why everyone is fixating on the 2025 numbers without even bothering to look up the 2026 numbers…
Curious who leaked this and why?
What's in it for them?
i'm curious what the two customers were
Meta is one
Seeing all the financial juggling AI companies do, it wouldn't surprise me if one of the customers is themselves lol
Where can I read this prospectus?
is the prospectus public yet? its not linked in the article afaict.
“ … according to its IPO prospectus seen by Reuters.”
Well now we know how/why opus 5.5 usage is so generous