upvote
> why should only profits matter?

In this case because it's not clear that anybody has actually figured out how to sell inference for more than it costs

reply
It's well know everyone is making great money on inference. The cost is training.

Whether GPT-5 was profitable to run depends on which profit margin you’re talking about. If we subtract the cost of compute from revenue to calculate the gross margin (on an accounting basis),2 it seems to be about 30% — lower than the norm for software companies (where 60-80% is typical) but still higher than many industries.

(They go on to point out that there are other costs that might mean they didn't break even on other costs - although I suspect these costs should be partially amortized over the whole GPT 5.x series, not just 5.0)

https://epochai.substack.com/p/can-ai-companies-become-profi...

https://martinalderson.com/posts/are-openai-and-anthropic-re... (with math working backwards from GPU capacity)

"Most of what we're building out at this point is the inference [...] We're profitable on inference. If we didn't pay for training, we'd be a very profitable company"

https://simonwillison.net/2025/Aug/17/sam-altman/

"There’s a bright spot, however. OpenAI has gotten more efficient at serving paying users: Its compute margin—the revenue left after subtracting the cost of running AI models for those customers—was roughly 70% in October, an increase from about 52% at the end of last year and roughly 35% in January 2024."

https://archive.is/OqIny#selection-1279.0-1279.305 (Note this is after having to pay higher spot rates for compute because of higher than expected demand)

reply
> It's well know everyone is making great money on inference.

That is not, in fact, "well known", but based entirely on the announcements of the inference providers themselves who also get very cagey when asked to show their work and at least look like they're soliciting a constant firehose of investment money simply to keep the lights on. In particular there's a troubling tendency to call revenue "recurring" before it actually, you know, recurs.

reply
> based entirely on the announcements of the inference providers themselves who also get very cagey when asked to show their work

I mean sure, it's self reported.

But the inference prices somewhere like Fireworks or TogetherAI charges is comparable to what Google/AWS/Azure charge for the same model an we know they aren't losing money - they have public accounts that show it, eg:

https://au.finance.yahoo.com/news/wall-street-resets-amazon-...

Fireworks’ gross margin—gross profit as a percentage of revenue—is roughly 50%, according to the same person

https://archive.is/Y26lA#selection-1249.65-1249.173

> In particular there's a troubling tendency to call revenue "recurring" before it actually, you know, recurs.

If someone has a subscription then yes that is pretty normal.

reply
> If someone has a subscription then yes that is pretty normal.

Not if you've substantively changed rate limits 3 times in the last 5 months while still counting those forecast revenues. In most industries that's called rug-pulling.

reply
It doesn’t matter how you call it. A recurring subscription on the books is a recurring subscription. Yes you can cancel anytime (how generous of them), it also doesn’t matter.
reply
not if your product is selling two dollars for one dollar and as soon as you'll start to charge more I'll switch to one of your twenty competitors

profit isn't a function of having a killer product, it's a function of having no competition

reply
And why do you think twenty competitors can stay competitive for years to come?

Industries always consolidate and winners emerge. SOTA LLMs look like a natural monopoly or duopoly to me because the cost to train the next model keeps going up such that it won't make sense for 20 competitors to compete at the very high end.

TSMC is a perfect example of this. Fab costs double every 4 years (Rock’s Law). It's almost impossible to compete against TSMC because no one has the customer base to generate enough revenue to build the next generation of fabs - except those who are propped up by governments such as Intel and Rapidus. Samsung is basically the SK government.

I don’t see how companies can catch OpenAI or Anthropic without the strong revenue growth.

reply
Google has already surpassed them both in all areas except coding. People on HN only look at benchmarks, but Gemini's multimodal understanding, things like identifying what a plant is, normal user use cases (other than chatting), integration with other tools, is much better.

It's believable that Meta, ByteDance, etc. can catch up too. It is not certain that scaling will meaningfully increase performance indefinitely, and if it stops soon, they surely will. Furthermore, other market conditions (US political instability) can enable even more labs, like Mistral, to serve as compelling alternatives.

Uber, TSMC, etc. have strong moats in the form of physical goods and factories. LLMs have nothing even remotely comparable. The main moat is in knowledge, which is easy to transfer between labs. Do you think all the money that goes into training a model goes into the actual final training run? No, it is mostly experiments and failed ideas, which do not have to be repeated by future labs and offshoots.

reply
> It is not certain that scaling will meaningfully increase performance indefinitely

It's certain that it won't. We've already hit diminishing returns.

reply
Google has completely caught OpenAI. Anthropic has a better coding model, but I'm sure Google is working on that too.
reply
> Anthropic has a better coding model

I’ll be polite and call this statement ‘a very debatable’ one.

reply
The barrier to replicating TSMC isn't just cost, it's supply chain, geopolitics, and talent.

Only one company on Earth can make the UV lithography machines TSMC buys for their highest end fabs, and they're not selling to anyone else.

The PRC tried to brute force this supply chain backed by the full might of the Party's blank check, all red tape cut, literally the best possible duplication scenario, and they failed.

reply
The PRC didn't fail, they haven't finished succeeding yet.
reply
They will succeed eventually since they have proof it’s possible and their plans span decades. I expect them to have working EUV in 10 years. Whether it’ll still be bleeding edge tech is a different question I dare not guess the answer to.
reply
>Industries always consolidate and winners emerge.

no, most industries just sell boring generic products, a few industries favor monopolists. Semiconductors are one of them but LLMs are also as far removed from that business as is physically possible.

TSMC makes the most complicated machines humans have ever built, a LLM requires a few dozen nerds, a power plant, a few thousand lines of python and chips. That's why if you're Elon Musk you could buy all of the above and train yourself an LLM in a month.

LLMs are comically simple pieces of software, they're just big. But anyone with a billion dollars can have one, they're all going to be commoditized and free in due time, like search. Copying a lithography machine is difficult, copying software is easy. that's why Google burrowed itself into email, and browsers, and your phone's OS. Problem for openai is they don't have any of that, there's already half a dozen companies that, for 99% of people, do what they do.

reply
no competition is a bit extreme. Limited competition yes due to competitive advantages.
reply