Again the issue is not is Anthropic an amazing growth story. Zero would debate that.

The issue is this is a probably $200-300B company pretending it’s a $2T company.

Add to that the insane amount of committed spend, a high interest rate environment which means they’ll pay a painful premium to patch cashflow gaps, and a likely “junk” bond rating (per Friday’s news) and that’s why folks are concerned.

It’s also just wrong to keep dismissing the paper losses from convertible notes as somehow not real losses. Once they are a public company this will be actual value taken away from existing public shareholders so they are legit losses not to be ignored. It’s like claiming equity based comp is free money… it’s not, you’re taking it out of the pockets of shareholders.

200-300B might be plausible depending on how market perceives the company

How can you confidently say that? At some point they oversaturate and growth stalls. Didn’t F100 have a major pullback already this year on AI spend?

SpaceX is down, Cerebras is down, etc.

I guess it applies to everything. Universal. Like LLMs.

What is "this" referring to here?

the difference in which side of this bet people take.

of course each stock price is many bets intersecting at the same time (with all their correlations of course)

stocks are claims on future cash flows, currency is claim on future purchasing power, but all can be though of as forms of money, where the prices are based on background information [ie. priors, beliefs, experience] and people's prediction of the future supply and demand.

and the general abstract definition of said bet is asking where are you in the information flow. if you know with sufficient certainty how the people afer you will evaluate some news then you can guess which side to take to make some money.

The SOTA US labs are one chinese open model away from being abandoned overnight by the masses. Ive argued before, but argue again here, that the current hardware shortages are the main thing keeping this from being a reality currently. If more people were able to run local models for a reasonable all-in cost, we would have a much higher velocity of iteration and improvement in the open source community. But its still too niche while people cant justify the cost, and so they continue to pay the premium for inference.

If a Chinese company released an open weight model that somehow was so superior to Anthropic and OpenAI then they can just take the model and deploy it, reverse engineer it, use it to improve their models, and/or do any number of things they currently do today which are of increasing value.

Most people don’t want to or have any interest in running or managing models locally.its not just about cost. The market for running models is infinitesimal in comparison to the market for managed products. That’s why everyone uses Office and not Open Office. Nobody is going to run some Chinese open weight model no matter how good it is on their critical banking infrastructure or something.

There will be space for lots of different solutions and many products and types of models including Chinese open-weight models and American ones too once we get around to making money on them.

The game isn’t just “cheapest open-weight model at all costs” which is what you are asserting here.

> The game isn’t just “cheapest open-weight model at all costs” which is what you are asserting here.

Decide: is the claim that open weight models will be more intelligent than Anthropic/OpenAI SOTA models, or that they will be cheaper?

> Most people don’t want to or have any interest in running or managing models locally

Most people are also not the primary customers of Anthropic! 5 companies make up 25% of Anthropic's revenue, and I bet there a ton of companies spending tens of $M/month on Anthropic/OpenAI that make up much of the rest. These are all medium/big companies that have the means for self-hosting models if they decide that's more convenient. The cost is of course not the whole story here, since Anthropic/OpenAI generally make better models and that allows using them in more situations, where perhaps other models would fail, but this is also a big cost for Anthropic/OpenAI to continue getting ahead, and the moment they stop they lose this advantage. Also in my experience self hosted open weight models are _fast_, which is _very_ nice for all those tasks where you don't need SOTA intelligence to get things done.

Where is their moat? Ultimately, where is their profit? answer? There is none at least there is none that will come close to whatever imaginary valuation Wall Street comes up with, in the best of times, while Wall Street may be able to carry one such company, but two within the same year?

No one has shown a network effect for models yet. Models are commodities. Anthropic etc will make the sort of profits one makes selling commodity goods (a couple of percent), plus a perhaps a bit of extra margin from being a few months ahead of open weight models (if that profit can justify the billions in training costs).

I do agree that local models are irrelevant. But local inference and open weights are different things. You can go on openrouter and see that there are dozens of companies selling open weight inference.

For one thing, one needn't run a model locally, that's a separate--but also very damning--issue.

For another, that's fine if Anthropic and OpenAI can do that, but they will have sunk a bunch of cost into their own training already and people wouldn't pay them what they need when there is always a vastly cheaper substitute that gets the job done.

When I argue the hardware cost, I should note that I think this is the limiting factor for _improvement_ of these models. But ultimately I dont think most users will self host in that world. They will still pay for inference, but it will be for one of a dozen commoditized vendors who are able to reap the benefits of inference at scale.

It's the same reason that open source software has been a commercial failure except for a few evangelists or whoever becomes a cool "thought leader" by releasing a new library and then gets invited to conferences to speak. Software was commoditized in many ways, yet all the big software vendors made more money than GOD because they took whatever commodity parts came to exist and monetized them.

I think the models can be improved with either recursive improvement (it would be faster with better hardware but if you can do it at all you just need time) and algorithmic developments. I don't know what those would be, but usually we see improvements to both over time. The algorithms will tend to be trade secrets as well.

The problem here is that let's say it's true that the Chinese are cutting off US companies at the knees. Well the US can just do the same thing back to the Chinese. We might have some money lost and stuff but the data centers will continue to be useful, as will all the infrastructure and energy grid buildout so it's not a total loss (I couldn't care less if accredited investors like myself lose money - you know the risk). Then we just steal Chinese models, do the open-weights thing, and all that stuff. So where does that leave us? Nobody makes any money? (Also where is Europe in all this?)

> I also dont think these need to be better than SOTA, just reach a "good enough" bar.

I agree with you in some respects but I'd like to digress here a bit. Good enough for a general purpose model isn't good enough. You want the best or at least what performs to your expectations. I don't even care about old models nor how much they cost (in my personal life and of which I will be a paying customer as will tens of millions of other people) I just always use the latest and best one unless something else triggers me to use a different one. It reminds me a bit about software updates. Folks around here are like, I'm not upgrading to macOS 17.2 and I'm staying on 17.3! Regular people? Not only can they not tell the difference they don't care - just give me the latest and best thing. period.

On the other hand you can have models that are OK at general purpose tasks but are perhaps specialized for certain applications (software or hardware). Those could wind up being on a spectrum from cheaper (and integrated with lower-quality products kind of like an Android Tv model) to very expensive and high-performant.

I don't see a world in which people are by and large sifting through well today I want to use Kimi 1.1 and well later today I'm going to switch over to Claude Lite 18.7 and dang it's not giving me what I want so I'll switch to whatever model. That's not how people operate - they don't want the cognitive overhead of trying to mess around with something that's good enough. They may have to suffer that based on their economic means, but it isn't going to be an expressed preference outside of very niche categories (like home-brew hackers playing with their toys).

Similarly this is just my opinion as well. I enjoy a good discussion here and I hope you are as well. Positive intents, even if my words my be curt, at times.

I have no issues with this thread : D But back to the discussion...

Again, I elided a lot of thoughts on the whole situation, but the other factor in my belief that this is an inevitability is that the SOTA labs churn too fast in both model behavior but also harness behavior. Normie businesses of the future are not going to hitch their wagon to a moving-target of both behavior and enforcement vs just pinning the equivalent of "opus 4.5" from a run of the mill inference provider. It just doesnt make business sense to be at the whim of a model update. And the entire perpetual motion machine aspect of the SOTA financial outlook _necessitates_ constant churn. If they dont release, they die.

> So where does that leave us? Nobody makes any money?

I think this leaves us with SOTA labs pushing hard for protectionist regulation to essentially require federal blessing to be an inference provider. I also think we will see a _ton_ of dark money poured into research to try and show things like theoretical smuggling/side-channel exfil attacks on models/model weights so that the SOTA companies can paint the _open models and model weights themselves_ as a bogeyman. If they can show that theres potentially latent attack vectors in those paths then that will then prevent all the "middle class" companies from being able to use them, especially if they need to adhere to eg SOC2/HIPAA certification etc

Also, because of the three hyenas of memory, the Chinese probably eventually will have control of the memory market too in the near 3 to 5 year future.

Win on what terms, or in what sense? We've seen this play out via Android versus iOS for example. The cheapest models are more prevalent, but Apple makes all the money. I'd be perfectly happy being the Apple in this case.

It's like you can go and buy a cheap Android phone. It technically does all the same things as the latest iPhone, but it does them worse. So Apple ends up making all the money and delivering the product that people go on to use to do all sorts of other economically advantageous things that make them even more money. Where does that leave the Android phone/model? In the dust, for the most part.

What you are arguing is that all progress on models (outside of government labs or internal corporate workings) will effectively stop because there won't be funding for it. That seems unlikely to me. You just need to play it out...

US companies/investors go and invest a trillion dollars or something on models. Someone "does stuff" and reverse engineers their model. They go out of business (you said so [1]).

So now what happens?

Well, the US/DARPA/universities/whatever (all countries with interest) will keep working on this stuff and there will be some funding for it but that's about it. Military applications, physics, things like that.

Chinese companies will also fail and it will not be possible to create new SOTA ones because they'll obviously just be reverse engineered too and given away for free or at super low pricing - whatever happens in America must happen in China too.

So who will build new models? There's nobody around to pay for it besides governments or maybe corporate labs. But corporations and governments will protect the models (obviously, as OpenAI and Anthropic do today to varying extents) and keep the weights and stuff closed source. They'll have some productive use for them even if it's not in the download Claude Code and have at it style of B2C/B2B we have today.

Or

Maybe the trillions of dollars of investment going into these companies actually makes sense and the threat of open-weight models isn't what you and others seem to be suggesting it is. Whatever fate we seem to want to assign to America also happens to China here. To the extent they are behind, they're still running the same exact race. If open-weight models undercut companies they'll also undercut Chinese companies.

Just as Open Office exists today, so does Microsoft Office. Microsoft just makes all the money. I'm not sure why some seem to think that they can use the open source comparison here but not also use the open source economics that we've seen play out.

I think as a productive exercise someone should run this argument and explain why Chinese companies won't also be eaten alive by this alleged race to the bottom and have all of their stuff reverse engineered and given away. How can they invest in new AI companies with SOTA aspirations? Whatever argument you want to make for the US applies to China as well, and we can just give away freedom models to all the countries just like supposedly they will.

[1] "The important distinction is SOTA US labs - ANT and OAI - who can not economically survive reverse engineering / fast following PRC models at PRC pricing."

>whatever happens in America must happen in China too.

No, because PRC literally could not do what US did... they simply didn't have access to enough compute to dump so much $$$ capitalization for SOTA race. On the other hand, on paper US WITH MORE compute and (currently) CHEAPER compute can put forth cheaper models to undercut PRC open models... but functionally they can't for the same reason cheap PRC operators can't win in US - geopolitics. Undermining PRC AI requires giving PRC cheap US AI on cheap US compute which is sanctioned. But ultimately that is entirely irrelevant to the fact that CHEAPER US models will undercut EXPENSIVE SOTA US models. The undercutting is not symmetrical. US companies can use cheap PRC hammers (or even cheaper US hammers) to break $$$ US SOTA. US who on paper can provide even cheaper hammers to PRC... but cannot sell those hammers... so PRC stuck using cheap hammers... and since PRC never had chance to spend $$$ on SOTA, there's no equivalent valuable things to break even if they had cheaper hammers.

> trillions of dollars of investment going into these companies actually makes sense

We know the context for which those $$$ investment makes sense... if SOTA held meaningful lead to sell high volume of $$$ tokens which was hedge 1-2+ years ago when years of $$$ commitments were made. But doesn't seem to hold anymore, ergo it no longer makes sense. Microsoft had decades to build ecosystem and moat, just like adobe it was, actually hard and high friction to move away. People are switching models all the time. SOTA AI only labs has much less moat than Microsoft or other incumbent tech with decades of overlapping capture mechanisms.

> I think as a productive exercise someone should run this argument and explain why Chinese companies won't also be eaten alive

The productive answer is THEY ARE ALSO BEING eaten alive.

The even more productive answer is PRC commodity AI doesn't have to win for US SOTA AI to lose.

There's just not much meat on PRC AI bones so they are losing billions instead 100s of billions / trillions. Like there is ongoing domestic AI involution irrespective of US SOTA. Literally the primary reason PRC AI was forced to go free was because DeepSeek is billionaire side project who priced his hobby as free which set economics for all other players. Being sanctioned/compute constrained = they figured out how to close SOTA gap on the cheap but also set ceiling on how much they can lose on race to bottom. In so far their SOTA aspirations, again refer to DeepSeek, he takes some couch change from hedge fund gains or investors, and throws it at smart people and hope it's enough. If its not then they'll probably stop too. What PRC not going to do is mobilize resources equivalent to several % of GDP concentrated in few entities for SOTA AI, at least not until their compute is sorted, in which case competitive exposure / risks becomes symmetrical. In the mean time, things are not symmetrical and arguments against economics of US SOTA AI does not apply to PRC commodity AI.

I think the fundamental disagreement is you believe that cheaper models can just break the state of the art. This seems false because if it were true then the SOTA can also stop developing and spending on SOTA and just use a cheaper model to build their own better models and leverage their existing compute and energy infrastructure (or even build it out for better and more capabilities later). Anything that is released the SOTA has access to, but it also has access to things that others don't. It's ahead by definition and it's not possible to come out ahead of it because the SOTA will just absorb whatever open-weight/open-source/whatever is released and add to its own lead.

> SOTA AI only labs has much less moat than Microsoft or other incumbent tech with decades of overlapping capture mechanisms.

And those companies are worth a lot of money, continue to execute at high levels, are systematically important, and are by any definition wildly successful. That's my point. You have to apply the open source style argument fairly and equally. If you're going to suggest the cheaper models (which follow this open source software argument) are going to come into being and do all this stuff you also have to acknowledge that they made no money in comparison to the Microsofts of the world and Microsoft won because it made all the money. Whatever open-weight stuff that gets released everybody gets to use it and so tech companies get to use it too and accelerate their advantages. This is what we've seen play out, why would AI be any different when you're using the same economic paradigm as your argument?

> Literally the primary reason PRC AI was forced to go free was because DeepSeek is billionaire side project who priced his hobby as free which set economics for all other players.

American billionaires can also "just set up a side project" and reset whatever economics exist. The crux of your argument is someone can do this and then SOTA isn't economically viable, but you can't escape that this has to be true for everyone, not just the US. If it's true for everyone you also then have to admit the SOTA can't progress because there's no viable economic model - anytime you make an advance in the SOTA someone "just sets up a side project" and/or steals your model or w/e.

The US SOTA losing just means the SOTA has to be equal for everyone or the US is ahead - you can't have it both ways. Progress is practically impossible because you can't make money and you can't obtain a lead so nobody will do any work since their work will just be immediately utilized by someone else.

If what you are proposing here is true, then China is permanently behind. There isn't an economic model to obtain a "lead", yet the American companies get to just take all the work that Chinese companies do (until they go bankrupt not making money) and add it to their lead.

> The productive answer is THEY ARE ALSO BEING eaten alive.

Well you can just run them out of business then. If they spend any money you just spend more until they can't make money and close up shop. Or is someone going to subsidize them indefinitely?

The economics of these arguments that the US SOTA companies are just going to go bankrupt because of cheaper models is something I just don't find compelling. I've been trying. I debate these things so I can learn, but I'm just not seeing a coherent argument on the topic.

I don't think all of these tech companies and investors are stupid. Greedy to a large extent and some will lose their hats? Absolutely. Stupid? I don't think so. Worst case is all the infra is still useful.

No their argument is that the existence of a worse quality cheaper product will eat the majority of the market for the SOTA company, making the SOTA company less profitable (not necessarily unprofitable) because the number of people who need/want the quality difference supplied by the SOTA product aren't enough to pay to sustain the extra resources required to supply the next iteration of the better product or get a respectable ROI on the current investment.

>SOTA can also stop developing and spending on SOTA and just use a cheaper model

Yes, and they will be forced to (if the market shapes out as above) but the point is that then the company won't be getting the returns that justify the valuation they are talking about in the IPO and will likely never return the amount of money that has already been invested in it even if it is a profitable company. i.e. If I spend $20,000,000 making a lemonade stand that is a successful best in class stand and it nets me $50 profit a day then it doesn't matter how that as far as lemonade stands go that it's a pretty amazing example. I am effectively never going to see that 20 million again let alone a real profit.

They're not arguing that these SOTA companies aren't better, they're arguing that there won't be a return on investment let alone justifing further investment and means everyone will be taking a huge haircut on what is already invested (with the possible exception of early investors who take the bag with the IPO when it inevitably succeeds based on people disagreeing with this evaluation and thinking it a good investment or simply because of index funds being required to buy in)

Cheaper models break financing SOTA pace, PRC budget fast follow currently is catching up. If SOTA slows/stalls slower SOTA pace = instead of PRC 6 months behind, they're 3 months behind, or period of overlapping parity before US SOTA grabs lead again for a few months. There may not be economic model to maintain durable lead, which SOTA economic doom spiral condition. Now if PRC fast follow couldn't close gap, i.e if PRC only fast following due to distillation and US SOTA can technically prevent PRC fast follow, and extend gap then maybe exorbitant SOTA spend sustainable. If not sustainable everyone moves to cheaper cycles. This an issue if SOTA cannot be done on the cheap, which also true for PRC, but PRC not doing expensive SOTA, they're doing cheap fast follow.

>why would AI be any different

It's not, someone going to grab big share of AI... but that someone has to be competitive enough for exquisite lock in, and frankly with enterprise reducing SOTA AI spend due to SOTA AI costs, current SOTA AI ain't it. See last paragraph for hopium.

> someone going to subsidize them indefinitely

Yes. Same way Alphabet/Meta can subsidize AI, China has their Baba and tech giants with alternate revenue streams to subsidize AI, even if not SOTA, and geopolitics will make sure US cannot run PRC AI out of business in PRC and vice versa. But geopolitics not stopping PRC companies running other PRC companies out of business, same with US companies running US companies out of business. In this case host countries tech giants with high margin businesses squeezing sole play AI labs. It's just US SOTA ANT/OAI getting murdered is massive economic contagion risk for US vs Zai/kimi not for PRC.

> find compelling

This really not complicated, last couple years, US SOTA labs signed 100s of billions of compute based on projected demand that they could monopolize enough demand to finance compute capitalization. This not stupid investor prior, because their only competitor, PRC was export controlled, compute constrained and shouldn't close gap. But PRC closed gap, made models good enough and architecture efficient / cheap enough so other players, including those in US, can capture large % of demand WHILE using fraction of compute. SOTA labs now looking at margin compression, losing share to good enough competitors and obligatory (now excessive) capitalization. This bad combo for current US SOTA economics. The numbers suggest this is what is happening. There is chance demand increases past projections and extra SOTA compute will be boon, but again, we are seeing most profitable SWE enterprise curtail not increase demand.

> Worst case is all the infra is still useful

IMO this actually best case for US medium term favour, but also another strike against CURRENT US SOTA. Reality is current SOTA is grossly overpaying for compute / AI premium. Even if OAI or ANT dies, their compute can be picked up for pennies on dollar and as long as PRC behind on compute there is potential for cheaper future US SOTA economics from cheaper hardware since crash will lead hardware price regress towards mean. But that boon is for the next guy.

Here's the US hopium: post current US SOTA crash, there is potentially curve where US has so much cheap discount compute, because current SOTA taking the haircut, and where PRC has not much compute, that US will have sustainable SOTA prospects, i.e. the world where US pays $1 for shovels instead of $10 and can buy $10000 worth of shovels while PRC can only buy $100. That's feasible after current exorbitant SOTA economics bleeds it's patients to death and the next US SOTA host get the transplants and lighter balance sheet, i.e. web 2.0 for SOTA AI. Incidentally more likely "winner" of AI... if I had to guess assuming incumbent FAANG doesn't just snatch up ANT/OAI if they explode - new player/entrant who are near SOTA and eventually SOTA but is priced more competitive than AI 1.0 SOTA, that enterprise will gladly pay for since not at current exorbitant rates that's causing enterprise to bail, but it will be fine because AI 2.0 SOTA can be sustained from 3T TAM instead of 30T TAM.

There's a world where capabilities narrow and Anthropic is stuck selling a commodity (not like they don't have any competition now). In this world it's not unbelievable that Anthropic would trade around or below 1x revenue no?

I think calling Anthropic a tech company could be like calling an electric utility a tech company.

The $2T comes from the implied odds (perhaps 10 or 20%) that we get near term AGI or RSI and AI becomes the entire economy, crossed with the odds that Anthropic gets there first and is the runaway winner (maybe 25%).

The trouble, in my view, is that today's revenue numbers are essentially irrelevant to the outcome that drives ~all of the expected value. The market has been pretending otherwise, but fundamentally we're looking at a globally top-10 public company that boils down to a binary call option, and the probability underlying that option is inscrutable. That's never happened before, and an inability to divine a fair valuation properly demands extreme volatility, especially in reaction to capabilities progress.

Finally (total speculation): I think Anthropic knows this. I think they're sitting on something with Fable 5.5 and the surprise outperformance of Opus 5.5 is a tell. They intend to release it just before the IPO and make some big claims underwritten by its performance.

At this point every churches should IPO and raise hundreds of billions from the implied odds that judgment day is near and Paradise becomes the entire economy, that wouldn't be crazier.

Anthropic simply has a better claim, better PR

most people have been already inoculated against that kind of doomsday predictions

(and also because of this there's a huge problem with solving slow moving diasters, like climate change)

Where have you seen a trillion dollar IPO from a cult?

In addition to:

> the probability underlying that option is inscrutable

The "positive" side outcomes are also pretty inscrutable imo: even if AI completely replaced labour, how does it trade against natural resources? Even if anthropic gets there first, might open weights arrive a few months later? If AI drives the value of labour to cents per hour, does that crater global demand?

For sure, the "anthropic stock is a ticket to heaven" scenario is in there somewhere, but even taking complete AGI victory as a given, it seems like a very narrow outcome.

On the third hand: if AGI victory is assumed, even if Anthropic isn't a great bet even under that assumption, where else to put your money? While the music plays, you have to dance.

FWIW Bitcoin had the same dynamics. Its valuation between 2011-2021 basically derived from a superposition of two futures: one in which it took over the entire global economy and became the dominant reserve currency, and one in which it failed and its value went to zero. The middle-ground was effectively excluded because currencies have strong network effects and a currency that is not dominant within a market quickly ends up unacceptable to merchants and worthless.

It's still somewhat in that grey zone, used as a form of digital gold by people who believe that the dominant reserve currency (USD) is going to zero as well as a trading medium by rogue nations, but its performance has been lackluster since 2021. It's currently up about 20% since then.

Entirely possible that AI follows a similar trajectory. People assume that it's either going to go bust or take over the economy, but people are really good at overlooking excluded middle grounds. Personally I think it'll settle into a space where it's very effective as a coding assistant, effective at creating propaganda and fake forum posts, and has the effect of killing Google Search (but not really replacing it), but doesn't fulfill all the hype around completely replacing the economy.

But a big part of the reason this has been true is that (mostly-software-based) tech companies haven't historically been spending significant billions of dollars every year in capex.

I think an open question is if SOTA labs should be priced like software companies have been. Software has been so valuable because the marginal cost approaches zero as distribution increases. Profit/employee at a company like meta prior to massive AI spending was measured in the millions. Meanwhile, the balance sheets of the labs look entirely different - remaining competitive requires continuous and astronomic capital costs, and there's no indication that this goes away. They are perennially being chased by competitors offering the same performance as models mere months old at fractions of the cost.

Their business to me, at least long term, looks a lot more like an airline, railroad, or other extremely capital-heavy industry with very competitive and low margins. None of those trade at 6+ p/s. Particularly on the buyer side, the switching cost to me as a high-AI-spending employee if a cheaper model offers compelling performance is less than deciding that I should ditch my preferred business travel airline.

I think that's why you're seeing so many calls from the leaders of those companies to regulate their competition away.

The big foundational labs are not like this. Every model they release takes a ton of data infrastructure and training (pre, train, post) compute. Then inference itself requires compute. Even if the labs sell tokens, each token embodies a lot of training and inference compute. This is a much more capital heavy business, so I can see revenue multiples being smaller than pure SaaS businesses. A model release bust (like Opus 5) ends up leading to big losses because of the necessary training compute spent on the model and the labor spent on training. A bust SaaS product ends up being wound down quickly, and only costs the labor spent on building it and the compute spent running it.

A good analogy is chip companies before all of the AI related compute pressure. These were largely low margin businesses because of the fixed capital cost associated with each new chip gen, architecture, process, etc

So I think the reason for the value gap is purely in the proprietary IP. The recipe for making a good LLM is more or less public, unless you buy into the "everyone just distils claude/gpt" angle. The recipe for making a good search engine is private.

Note the huge difference in strategy and outcome!

Google published essentially nothing on how to build a search engine. Huge R&D spend, zero publications, intense culture of secrecy, extremely low attrition to startups or competitors. They did publish some papers on big data systems but only on tech that was many years old - a typical Google "research" paper pre-2015 would announce that the system it described had been running at scale in production for years already. Google also never faced any kind of split or mutiny in which lots of people with the core secrets left the firm to set up a competitor.

With AI however this changed and they published all their research immediately, also their best employees constantly ran away to join startups. The result was predictable: smaller competitors deployed their ideas faster and better than they did, and now Google has no moat anymore.

OpenAI took the classic Google approach and stopped publishing R&D after developing GPT but this didn't help them build margins because:

1. Apparently the core transformer idea is powerful enough that breakthroughs after that didn't matter much. Nobody has been able to develop some killer edge that leaves competitors unable to catch up. Most improvements to transformers have been about cost reduction, not capability improvement.

2. They had a huge split when the Anthropic guys walked out over safety, taking the knowledge of how to build a frontier LLM with them. Result was a peer competitor from day one, whereas Google had at least six years to build edge before anyone realized web search mattered.

Result: Google-style hardware costs but without any secret IP that could generate revenue commensurate with those costs.

Because we're just waiting for the world to find out that they can spend 1/10th and get virtually the same results.

Open source models are now 1/100th in price per task of US sota from this summer...

With the price of AI rapidly falling to the price of compute, with models therefore basically having zero value, and with Anthropic having rented all its compute from others then there’s little evidence that such high profits are in the future.

Anthropic has a great product but is for sure a bad investment.

If this is the new normal, that's great news. Revenue growing at 47% per quarter will 10x in 6 quarters. Even if you think it should be a $200B company today, it'll be a $2T company in 18 months, assuming the same revenue multiple.

If you think this is the start of a rapid decline to 0% or negative growth rate, then it's obviously bad news.

The real answer is probably somewhere in the middle -- but exactly where in the middle you think it will fall depends more on your own priors than anything you can read from the data.

What do you expect for FY2027 revenue to confidently conclude this? not saying you're wrong, but considering the dramatic changes in their revenue you seem strangely very certain that you know where it will land.

It does not make sense but it is what they say

The only way to get value from acquiring other companies is to buy them when they're distressed.

Take a look at how OpenAI has shaken up higher level math.

Frontier AI labs have a Swiss army knife of a tool that can do as well or better than every other company or collective in every industry.

Say you have a magic hotpot that can deliver any dish you want and each dish it delivers can be as good or better than any restaurant in your area of business. You basically cannot be beaten in the food business. Your competition can only cry about their cooking being more artisanal or whatever but at the end of the day convenience and cost will triumph.

Now say you have a magic pen for art, a magic jukebox, a magic text generator for authorship, and so on. You get the idea why I don't think Anthropic is wrong.

So I think they can compete in many industries (and I don't like it), but not all of them for sure. They can only compete in all industries with infinite money, and I don't think it would be good for most people.

The places where American society needs costs cut or needs an increase in labor - healthcare immediately comes to mind - are places where you'd be asking Anthropic (or any other AI player) to take on massive amounts of liability.

If you screw up someone's drawing, you regenerate the drawing and charge the hack/fraud for the privilege. If you screw up a blood bank order, you pay the survivors a few million dollars. Since you've created a system that can scale blood bank order up-screwing, you'll be doing that a lot. There's not profit in paying out medical torts.

Given that SV-based companies work to dismiss liability even more than most American companies do (which is saying a lot), I don't see that working well for them.

They probably asked Claude. Sounds like a normal LLM hallucination.

Should have followed it up with a second prompt to get the real answer ... "You're absolutely right. valuation is only $20b"

Also, anthropic has no real moat, LLMs are a commodity.

They reported a loss here because they sold notes convertible into shares and their valuation went up. Normally shareholders do not lose money when the stock price goes up.

It's weird that they had to report it like this because they are technically derivatives, and if they had just given out actual shares (as with stock-based comp), they wouldn't. Yes, giving people stock is a real cost, but what happens to the price afterwards is not, and when the price goes up shareholders are definitely making money, not losing it.

Either way, they did report it. It is up to you to decide how you interpret that information.

> Okay so Q3 ARR at $67.8bn versus $46.2bn for Q2, +47% QoQ. Turns out tokenmaxxing's end over the summer slowed revenue growth versus an absolutely torrid 100%+ growth rate in Q1/Q2 but 46% sequential is still eye-popping

> Operating margins (GAAP) went from 8% to -3$, with a -28.5% QoQ incremental operating margin. $SPCX compute coming online is the reason.

> Bottom line after seeing this Q3 update: the basic takeaways from the original S-1 don't change, but fears that they were excluding Q3 because it was a disaster were definitely wrong. Do with this information what you will.

It's also worth noting that Q3 only has one week of Opus 5.5 (9/22 - 9/30) which may have changed the economics for Q4.

That's a very generous reading. Anthropic's entire growth narrative is the rapid growth over the previous 12 months, from near-enough zero to $100bn+ this year. Now we see that actually growth is levelling off and that their revenue was ~$5bn/month over the last quarter, which is barely over their compute commitments, and nothing close to the $100bn+ leaked.

I guess if people thought Q3 would show negative growth rate, sure, it isn't a disaster, but their growth rate slowing so much is surely very bad and matches all the concerns commenters had.

And this is Anthropic! Anthropic are the financial success compared to OpenAI. OpenAI are doing much worse, so this does not bode well for OpenAI either.

Really, you're talking about expectations ("expected") of the expected ("price") trajectory (actual business performance).

If the above comment was intending to say "chart go up steeper make price higher, chart go up less steep make price lower", then no dispute there. But that carries virtually zero predictive information about whether the business will achieve "escape velocity" (whatever that means).

You absolutely cannot predict Anthropic's fate based on their growth rate slowing from Q2 to Q3. Obviously.

You can predict the resting place of a projectile with information like this. You cannot do so with a business, especially one as dynamic as Anthropic in an environment as dynamic as today's.

That reason possibly being to grant legitimacy to an otherwise mostly illegitimate and equally mostly fictional branch of the economy (that whole MBA cluster).

The main difference I would concede is that gravity, mass, forces, awind drag etc are much easier to model.

Businesses can have stalled growth, degrading margins, etc. etc. and then simply... reverse it. They can raise more money. Their competitors can get regulated out of existence. They can do a literal infinite number of things that cannot be factored into a model like "a projectile."

That's why the projectile analogy is stupid as fuck. It suggests you can presuppose the conclusion by presupposing the conclusion.

No one said it literally is a projectile. It's an analogy and model. The similarities are in the present tense maths. It's presumed and beyond having to be stated that businesses exist in a complex set of variables all of which could change and might not be as predictable as the valuation maths.

You're not wrong, you're just not adding anything to the conversation that any but the most uneducated do not already know and acknowledge.

It implies that you can know Anthropic's outcome from this slice of data. That is literally not true in any way at all.

Like it's not even remotely or analogously or roughly true.

Anthropic behaves like a high-growth business with a quarter of slowing growth. It does not behave like a baseball. Any conclusions you draw from analyzing it as a baseball will be de facto faulty because the two systems are nothing like one another. The most basic assumption of projectile analysis (that energy is not added in-flight) is the opposite of the most basic assumption of a business (that they are actively profit-seeking).

The analogy adds nothing except ambiguity at best and active falsehoods at worst.

No it doesn't. All I said was that given how little we have to go on and how early in the process we are, a minor 2nd or 3rd derivative slowdown isn't a positive signal. All indications are they they need additional orders of magnitude in revenue fairly soon.

I also never said that it wasn't self-propelled. I was starting to write about waiting to see if another stage of propulsion comes online but I deleted it.

It's saying "like a projectile, you can predict the outcome based on just a sliver of data"

That is literally not true.

Most businesses are stacks of S curves (except when they're not).

And in 10 years it will be worth the entire Earth's economy.

Well that’s the point. They’re not continuing at that rate of growth.

47% quarterly growth is still 4.5x annual growth, which would put them at $300+ billion revenue in a year. This is still unprecedented growth for a company of this size.

The negative perspective depends on assuming that the 47% quarterly growth is going to continue dropping soon. And it is going to drop, since it's only a bit more sustainable than last year's growth. The question is how many hundreds of billions they're making a year when it levels off.

I really don't understand how could it be otherwise? They're planning to IP at a valuation representing 5% of the US GDP and, unlike NVidia, they are neither selling shovels in a goldrush nor a business with decades-old proven track of records.

Are people really going to buy ANT thinking it'll 10x to 50% of the US GDP?

Aren't there better stocks to YOLO on if you think you could maybe 10x? And if you want something safe instead of YOLO'ing, aren't there plenty of companies much safer and much more boring to invest in than something IPO'ing at 5% of the US GDP?

Isn't the entire Anthropic narrative based on infinite growth? How is infinite growth sustainable?

This is why any [consumer] investment in ANT is bad. If ANT goes 10x then congrats, we [the citizens] are in a crisis. If ANT goes 1/10x then congrats, holders of ANT are in a crisis.

https://x.com/billylo/status/2117276235852345425?a=4-pjgf1Ox...

Some people just start out with wanting LLMs/AI/Anthropic to fail or be scams and consistently try to fit any new evidence pegs into those unchanging hypothesis-shaped holes. This group ideologically despised LLMs from the start and needs them to be fake.

We're nowhere close to the ceiling on LLMs alone, let alone AI in general. What about when they can run entire firms themselves? Like, this is just assuming Opus 4.6 is the best we could ever get, in which case, yeah, the notion of AI is simply a dead-end technology. But obviously it's not going to be the best we could ever get.

That would mean the whole idea of free markets will be rendered obsolete, instantly destroying the idea that investing in Anthropic is sensible.

Why assume that?

So much money have been given to that handful of companies and to that field, that it's pretty much guaranteed to plateau (it's not some small but growing market with lots of room to grow. It's already taking as investment a huge chunk of the economy, and the returns don't seem to justify it).

Most of it is a pre-IPO hype and a circular financing loop between Nvidia, Oracle, Amazon, and the AI players anyway, so the real returns are even worse.

As for the rest of your post: bizarre to see such ignorant slopulist drivel on HN of all places. Is this a YouTube comment section or an attempt at an intellectual message board?

Thank you for your insightful argument.

https://xkcd.com/605/

Now, they need to unearth new use cases and saturate them to keep the momentum going. Its unclear if they have found these opportunities and can turn them into sustained growth.

Within large tech companies the era of "spend the most money on the smartest model running on its maximum effort, always" is dead.

This is specifically about the end of tokenmaxxing and how much worse the growth rate would look after the Q3 numbers came out.

(I do think moats are shallow, and I personally switch away from Anthropic every few months for a while, but I've been coming back. All software moats have been made more shallow now, affecting the entire investment landscape.)

Anthropic is working hard to create a vertically integrated stack from model to harness to productivity apps. That may or may not work, but it’s clear they see models as a value add for integration and not a product that can be a moat in themselves.

Codex also wins by consuming a lot less memory.

I’m not convinced it’s durable or that Anthropic will be the ones to change the document habits people have decades in… but it is a mistake to only see these companies in terms of models and coding/chat harnesses. There is something a lot bigger there.

https://openrouter.ai/compare/anthropic/claude-opus-5.5/deep...

By the way, we get way more value from our Anthropic subscriptions than any Chinese subscription.

Your posts are irrelevant.

Evidence? I was told by a lot of people that token use was going to start to slow down 5 months ago because of "the end of tokenmaxing."

openrouter still showing weekly token usage higher than ever: https://openrouter.ai/rankings

No, a lot of people like Gary Marcus said it had specifically already ended in May/June. Why is this time different?

>Tokenmaxxing is over. https://finance.yahoo.com/sectors/technology/articles/tokenm...

>Tokenmaxxing is dead, long live valuemaxxing https://www.ibm.com/think/insights/tokenmaxxing-dead-long-li...

>Token Maxxing, April 2026-May 2026, RIP https://x.com/GaryMarcus/status/2057610132742156603?lang=en

There is an unbelievably enormous difference in 100% vs 47% -> and when you look at the 2nd derivative this is REALLY REALLY REALLY bad.

I mean, the growth is astounding and incredible. It does not bode particularly well for a ~$2T valuation.

An average random dev here: I switched to Opus 5.5 and forgot about all other (paid) models.

This should help build some intuition and establish a baseline for what the $2T valuation means.

If you buy at $2T Anthropic needs to be worth about $3T by 2030 for a normal market return. That's ~25x P/E and ~$120B/year of profit.

That could be something like $400B of revenue at 30% net margins, $300B at 40%, or $600B at 20%.

It would take 1.7x revenue growth a year to get from today’s $68B run-rate to $400B by 2030. Anthropic's current growth rate is ~4.7x a year. Anthropic doesn't need to maintain that, they basically just need to prevent it from absolutely collapsing and hope margins hold up against open weight models.

You could look at this and say "yeah that's plausible" or "I highly doubt they achieve that" but it's hyperbolic to say its actually impossible they can meet this valuation, definitive proof of bubble, cant be worth this unless they can create God, etc.

There are so many variables involved that are outside their control. If GPU/RAM demand dries up (or supply increases), for example, local models are suddenly a threat in addition to hosted open weight models.

The quality there is already good enough. I've been 100% local for personal projects since Qwen3.8 Flash dropped. It's not as nice an experience as Claude I use for work, but it gets the job done and I fully control my data and can do whatever I want with it.

Maybe they tried too hard to get good Q2 numbers, making revenue happen earlier and expenses later, and that now makes Q3 look worse? Either way, this does not look great

And they say that this will likely continue to grow until 2027 (when the long term commitment compute comes online, which comes with opposite problem .. will there still be enough demand)

But are they still a 1.5T USD company if their growth is already slowing this much?

Additionally, the Anthropic S-1 does indeed show scalable profitable growth. This is perhaps the most shocking thing in the S-1, to me.

A better way to phrase your thinking might be that revenue is an indicator of demand.

The issue here is that they've not really tested what happens to demand when the prices moves to where it needs to be to cover the cost of keeping the lights on.

>People should really look at cash flow statements instead of just earnings. Anthropic published one and their cash flow was positive for the first half of this year.

All of that isn't very newsworthy though. We knew all that from the Q2 numbers. What is interesting is that reinvesting all their profits to make the number go up in the next quarter didn't work as well this time as it did all the other times

They stopped when the free packs stopped coming and they needed to finance their own.

Right now, I could start a business selling gold bars for $1 a piece. There would be more people interested in that than Anthropic.

ANT also has positive gross margins, so they aren't selling gold bars for $1.

I don't understand why people on HN treat ECON101 like some fringe theory every time this is discussed. This is basic stuff.

They have yet to release a model that has made more in revenue than it cost to develop...and each new generation costs more to develop than the preceding generation. Costs increases outpace revenue increases with every generation, across the industry, and LLM-based cash flows are highly negative even at Google and Microsoft.

Customer use of LLMs is decreasing after quarters of growth. Execs were all in on pretending LLMs were the future until the WSJ/Stanford report revealed what everyone outside of tech already knew: LLM use has not correlated with any measurable productivity growth in any non-programming industry. (And arguably, the same is true in tech as well: LOC has increased but quality has dropped, so the net change to productivity is de minimis.)

Apparently you’ve missed all of Ed Zitron’s reporting?

So he can get all of his conclusions and predictions wrong based on his numbers and analysis, and then you think his value is in his numbers and analysis?

Since the start of the year the market has shifted from 50/50 split (closed v open weights) to 30/70 (70% of usage they see through their platform - is now open weight)

DeepSeek V4.1 Flash 1017 sees 5x the weekly token usage through OpenRouter than the closest closed weight model (GPT-6 Luna)

Even though OpenRouter is only one limited view of the market - those trends should be worrying to potential investors.

OpenRouter could so 90/10 split and it still wouldn't mean open models are beating (in usage) closed models.

That makes any loss or risk reasonable.

It's great propaganda, so great that they have themselves started believing it with certainty. This is one of the self owns that brought down the USSR :)

Of course they could be correct too, time will tell.

Then if all goes well next year OpenAI will scam retail investors by claiming they will have the only A.I. worth anything.

https://searchengineland.com/guide/facebook-users

Consider: if you had built a hyperintelligent stock-trading AI, would you:

- Charge $0.70/M tokens for other people to ask the bot what stocks to buy.

- Get rich following the bot's trade advice and never tell anyone about your bot.

The fun thing about open markets is the market will punish you for this with second order effects and it's in your best interest to get some whales on the bandwagon.

Pretty much every technology in history is available at several orders of magnitude lower cost than the first few years of it being invented.

AKA eventually everything runs on a potato.

Not today's raspberry pi, but the equivalent of it absolutely. I don't work in it, but have a number of friends who are involved in LLM development - the "breakthrough" optimizations are obvious work in hindsight. We are still in the very early innings of optimizing both hardware and software techniques. The recent market imbalance has blurred the idea that hardware simultaneously gets both cheaper and much better, but no one serious expects that to stop on a longer timeline.

Going beyond the "my watch has more computation power than NASA at the moon landing" - speech-to-text used to be unreliable tech that needed datacenters backing it as recently as the early iphone, and now runs in near-realtime on smartphones that cost about the same. A ~$50 processor running in my closet does realtime person/vehicle/package detection on my front door camera reliably.

https://huggingface.co/microsoft/bitnet-b1.58-2B-4T#evaluati...

Which LLM use cases can be solved with compact decision trees is an awesome open problem.

If we're talking web servers, sure. Video games? For the most part. Numerical high performance computing? ehhhhhhhhh i'm not sure history bears that out

Any numerical high performance computing from 1995 will run on a 100$ android phone.

https://en.wikipedia.org/wiki/Cray_T3E

And if this was the fastest, the average ones were significantly smaller.

What you're missing is that the things you mentioned sometimes took months to run in the fastest computers thirty years ago, but you are mentally expecting them to work quicker now.

The open weight models right now are nowhere close to ..say.. Fable 5.1

They could catch up if Anthropic reaches some sort of a plateau and they manage to keep improving, but there's no reason to be certain about that happening.

Which was probably the right move. The S-1 didn't seem to really change the overarching economic and financial narratives about Anthropic (it seems like whatever people already felt before the S-1, they felt it even more after looking at it).

So I don't think there's going to be any problem with boosters metabolizing the Q3 numbers.

(Idk, maybe this was their plan from the jump.)

It says "You're absolutely right!".

So they did release the reports.

Tokens themselves are already a bit of a commodity, and the margins are gonna be fierce on these as there is not a whole lot of lock in.

Would Anthropic and OpenAI position themselves using their extra tools, their ai agents, presentations, docs. Are they nice; yes. However, that doesn't make a 2T company. It reminds me a bit of Dropbox, a truly great product, however, they're differentiation is simply not there. I think the momentum is there for the IPO to go decent, but it feels like the company is on such unstable ground that the fallout could be catastrophic.

Google's strategy right now seems great.

Owning the best search index and being able to retrieve content immediately straight out of their index seems like such a massive competitive advantage.

FWIW, from my prompt last night: https://chatgpt.com/share/6ae8d441-8c44-83e9-980c-b14a2ed846...

Consolidated financials are on page 23, quarterly numbers on page 160

In Q3, I think they struggled with the government ban on Mythos/Fable, Opus 5 not being so good, OpenAI models being more token efficient, and they decided to dedicate a higher percentage of compute to training.

It’s not possible at all that they had to cut prices to hold market share as folks switched some of their usage towards open models.

Utterly cringe takes.

and add'l discussion: https://news.ycombinator.com/item?id=50338510

Claude has been pushed into every software org possible already by force. Companies will evaluate the actual benefits in 2027 and usage will go down.

There is no other product. It remains to be seen how long investment banks and influencers can spin the narrative.

Too much sensationalist reports out there, cherry picked by all sides.

Look, I've seen and I understand the arguments that the GAAP income adjustment for this kind of thing is fake, but a number of them seemed to take it for granted that it was a "one-time" adjustment. If it's going to keep happening I really feel like we ought to insist on a more comprehensive argument. The accounting principle must be generally accepted for a reason: what shenanigans could someone get up to by reporting net income without this adjustment, and how do we know Anthropic is not getting up to those shenanigans?

We aren’t shareholders, though, so why should we care?

They're going through an IPO, so unless this is a one-off thing, people reading this thread will become shareholders, either directly or through their pension pots.

So while yes, this does look a lot like shareholder dilution, that doesn't really satisfy the question of why it's not treated that way under GAAP. Perhaps GAAP is just wrong, or perhaps there's some possible tricks which Anthropic is provably not engaging in. What it's definitely not is a one-time adjustment now that it's happened in two consecutive quarters.

The reason it happens in multiple quarters is likely that Anthropic's valuation has increased in multiple quarters. Every time Anthropic's valuation increases, the convertible bonds are worth more money. Other shareholders have "lost" money compared to what would have happened if Anthropic had sold regular bonds instead of convertible bonds. Except that no, the value of their shares is increasing too. It's actually good for them.

That's how equity works. If someone owns 1% of a company and its valuation doubles, their share is worth twice as much and so is yours and we wouldn't even normally consider it to be diluting the other shareholders.

So the question is whether convertible bonds are worth more than regular bonds and the answer is, yes, absolutely, but they are partially equity, and it's the equity part that's increasing when the company's valuation is increasing.

And if the company's valuation went down then the value of the convertible bonds would decrease. Maybe that would be considered a profit?

It is like someone who loses repeatedly at Roshambo while saying: "Okay, then. Best two out of three- three out of five- four out of seven- five out of nine..."

[0] https://edition.cnn.com/2026/10/30/business/ai-stocks-bonds-...

Obviously the investors who bought in at the bottom did the best :-) Most people who are crying 'AI bubble' predicted 8 of the last 2 bubbles.

My personal opinion is ANT IPO is like the netscape IPO of 1995, its the start of the cycle not the end. Things are going to get really crazy over the next 3-5 years, the last year was just a warmup. I don't think people realize how much credit fiat dollars have been and can be created, and how $1T companies will be commonplace. I think humans have a mental block, where we think dollars are like rulers or other fixed measuring devices, so you can look at the same number at different periods of time and compare. credit backed currency completely warps our sense of scale when it comes to measuring things in dollars.

It's not the same thing as early NVidia / Amazon / Google IPO: these companies didn't IPO at a value immediately placing them in the top 10 (by companies market cap) and certainly didn't IPO at values representing 5% of the US GDP.

Can it x2? x3? x4? Maybe.

Can it x100 and represent five times the US GDP? I don't think so (I know, I know, the US GDP would grow too).

I know ANT ain't SPCX, but look at what happened so far to SPCX's valuation.

See: https://news.ycombinator.com/item?id=50157976 and AI Coding Agent Security Benchmark https://www.endorlabs.com/research/ai-code-security-benchmar...

https://www-cdn.anthropic.com/daad4360a8bdc707f8b22e3e745796...