5 ms·
>3. We're massively overusing SOTA models. As long as you're on a subsidized subscription, you can use Claude Opus 4.8 high to write blog article meta descripti
by LUmBULtERA 3mo ago
>3. We're massively overusing SOTA models. As long as you're on a subsidized subscription, you can use Claude Opus 4.8 high to write blog article meta descriptions. If you paid by token, you wouldn't do that.
This idea that the subscriptions are subsidized is repeated over and over, but I've never seen any proof of this. It seems to be entirely based on the inferred API cost the subscription usage could give you, but there are a lot of assumptions needed for that to follow.
- erfgh 3mo agoThey are subsidized by the huge losses incurred by the AI companies.
- chilmers 3mo agoOnly if those losses are coming from subscriptions, instead of capex and training, which is not at all clear.
- Hamuko 3mo agoI don't understand this argument. How does it make the subscription any less subsidised if the losses are only because developing the product is just so darn expensive? Feels like arguing that it's not clear if Bugatti's losses came from selling the Veyron instead of designing and developing the Veyron.
- fragmede 3mo agoThe equivalent is when Amazon was running a loss because they were spending all their money on building warehouses. It exactly make sense, but that's the argument.
- smokedetector1 3mo agothis argument assumes that capex and training costs will go down over time. but theyll have to keep up with one another and stay on top of latest knowledge so Im not sure if thats true
- Yizahi 3mo ago"Our 2015 car models are totally profitable if we will just stop making new cars and continue producing only 2015 models for the next decade."
- winternewt 3mo agoFrom the article: > What is happening here is that leading AI labs are charging not only for inference but also for research in model architecture, training data collection and curation, model training cost (which can be tens or even hundreds of millions of dollars), paying their employees and recovering the marketing costs. That's what's being subsidized.
- yreg 3mo agoYou are saying it as if those costs were not necessary to provide the service.
- LUmBULtERA 3mo agoOpenAI inference revenue exceeds its cost of inference by a good margin in 2025 (https://cdn.arstechnica.net/wp-content/uploads/2026/06/openaifinance.001.png https://cdn.arstechnica.net/wp-content/uploads/2026/06/opena...)
- jijijijij 3mo agoGreat, but that's only a part of operational costs. A craftsman's revenue may exceed the electricity bill for the power drill, doesn't mean the business is sustainable.
- LUmBULtERA 3mo agoDay 2 the craftsman has not made up for the investment/loss of their equipment. Not a useful example.
- jijijijij 3mo agoSorry, I don't understand what you are trying to say.
- LUmBULtERA 3mo ago
- usef- 3mo agoAnthropic have claimed they expect their first profitable quarter this year. As far as we can infer their current API prices have decent margins.
- Yizahi 3mo agoAnyone can claim they are profitable, simply by reclassifying their expenses as some other thing or shuffling them to separate corporate structure. Until we will real financial audit, the CEOs claims are just a hot air.
- usef- 3mo agoOpenAI's leaked documents also said OpenAI was profitable on inference. The small resellers of open models have nowhere near the resources to optimise their models or inference and yet usually have a lower cost, why wouldn't the big labs?
- rightbyte 3mo agoThe leak really doesn't say that. We have no way of knowing what posts are in the different categories.
- Yizahi 3mo agoThat is exactly the document I've been thinking about while writing top comment :) . Oh, our "cost of revenue" is smaller than the revenue, we are so profitable guys! If we just don't count our marketing expenses, our administrative expenses, and our unspecified losses from operations to the tune of x3 times higher than our revenue. But if we don't count them we are totally profitable guys! If we will just stop all RnD and also stop paying our salaries, support and marketing departments (and also cut 20 billion of operational losses too, whatever that is) we will earn soo soo much monies guys!
- LUmBULtERA 3mo agoFair to include A&G, but their marketing and training cost is about generating future growth. People say that the subscriptions are highly subsidized as if its fact -- it's certainly not a well-established fact. These people simply don't know the truth one way or another, but portray their not well supported opinion as fact.
- LUmBULtERA 3mo agoData from OpenAI shows their 2025 inference revenue exceeded their cost of inference by a good margin (https://cdn.arstechnica.net/wp-content/uploads/2026/06/openaifinance.001.png https://cdn.arstechnica.net/wp-content/uploads/2026/06/opena...). Saying this is being subsidized is like saying any investment in future productive assets is "subsidized".
- player1234 3mo ago[dead]
- heyoni 3mo agoWhat assumptions are needed for inferring cost based on api pricing?
- gck1 3mo agoThat API pricing isn't massively inflated.
- LUmBULtERA 3mo agoThe API inference cost to customers is not the actual cost of providing inference, and the cost of providing API inference need not be the cost of providing subscriber inference.
- xboxnolifes 3mo agoAPI prices are consumer prices, not costs.
- oarsinsync 3mo ago> This idea that the subscriptions are subsidized is repeated over and over, but I've never seen any proof of this. It seems to be entirely based on the inferred API cost the subscription usage could give you, but there are a lot of assumptions needed for that to follow. My claude code environment shows me cost per token used in that session, according to API costs. It regularly exceeds $200. I pay $200 a month for my claude subscription. That's fairly obviously subsidised, unless you genuinely believe their unit costs are 100x less than what they're charging.
- qtk8 3mo agoBut they also determine the token prices. What you describe could also be true if they take a 5x profit margin on api tokens and 2x margin on subscriptions.
- hparadiz 3mo agoThat's what they want to charge you. Not the actual cost. The actual cost is a gpu that's probably already paid off and about $2 of electricity
- t-writescode 3mo agoMost prices, like GPUs, are amortized over several years, when doing the calculus. Maybe they're already paid off, maybe they aren't. I would lean toward "aren't".
- hparadiz 3mo ago[dead]
- LUmBULtERA 3mo agoThe API inference cost to customers is not the actual cost of providing inference, and the cost of providing API inference need not be the cost of providing subscriber inference.
- 3mo ago
- JBlue42 3mo agoBecause no one knows what the true cost is, especially with how all the money is circular. For example, Microsoft 365 Copilot. A company might get a significant discount on the price per user for that SKU. What does that translate to? OpenAI is the actual brain behind it. So who ends up getting the money at the end? https://blogs.microsoft.com/blog/2026/04/27/the-next-phase-of-the-microsoft-openai-partnership/ https://blogs.microsoft.com/blog/2026/04/27/the-next-phase-o... OpenAI needs Microsoft's infra for training and market penetration. MS needs something AI to slap onto their products until they develop their own in-house. Now, whether it's actually necessary or not for an enterprise is a question. There's a lot of FOMO spending. Maybe 5-10% of your workforce are programmers that need great AI. The rest of the admin, finance, other people? TBD.