"Do LLMs make bad code: yes all the time (at the moment zero clue about good architecture). Are they still useful: yes, extremely so."
Well, lets see how all the economics will play out. LLMs might be really useful, but as far as I can see all the AI companies are not making money on inference alone. We might be hitting plateau in capabilities with money being raised on vision of being this godlike tech that will change the world completely. Sooner or later the costs will have to meet the reality.
> but as far as I can see all the AI companies are not making money on inference alone
The numbers aren’t public, but from what companies have indicated it seems inference itself would be profitable if you could exclude all of the R&D and training costs.
But this debate about startups losing money happens endlessly with every new startup cycle. Everyone forgets that losing money is an expected operating mode for a high growth startup. The models and hardware continue to improve. There is so much investment money accelerating this process that we have plenty of runway to continue improving before companies have to switch to full profit focus mode.
But even if we ignore that fact and assume they had to switch to profit mode tomorrow, LLM plans are currently so cheap that even a doubling or tripling isn’t going to be a problem. So what if the monthly plans start at $40 instead of $20 and the high usage plans go from $200 to $400 or even $600? The people using these for their jobs paying $10K or more per month can absorb that.
That’s not going to happen, though. If all model progress stopped right now the companies would still be capturing cheaper compute as data center buildouts were completed and next generation compute hardware was released.
I see these predictions as the current equivalent of all of the predictions that Uber was going to collapse when the VC money ran out. Instead, Uber quietly settled into steady operation, prices went up a little bit, and people still use Uber a lot. Uber did this without the constant hardware and model improvements that LLM companies benefit from.
> if you could exclude all of the R&D and training costs
LLMs have a short shelf-life. They don't know anything past the day they're trained. It's possible to feed or fine-tune them a bit of updated data but its world knowledge and views are firmly stuck in the past. It's not just news - they'll also trip up on new syntax introduced in the latest version of a programming language.
They could save on R&D but I expect training costs will be recurring regardless of advancements in capability.
If the tech plateaus today, LLM plans will go to $60-80/mo, Chinese-hosted chinese models will be banned (national security will be the given reason), and the AI companies will be making ungodly money.
I'm not gonna dig out the math again, but if AI usage follows the popularity path of cell phone usage (which seems to be the case), then trillions invested has a ROI of 5-7 years. Not bad at all.
OpenAI would still lose money if the basic subscriptions were costing $500 and they had the same amount of subscribers as right now. There's not a single model shop who's ever making any money, let alone ungodly amounts.
These costs you are referencing are training/R&D costs. Take those largely away, and you are left with inference costs, which are dirt cheap.
Now you have a world of people who have become accustomed to using AI for tons of different things, and the enshittification starts ramping up, and you find out how much people are willing to pay for their ChatGPT therapist.
Private equity will swoop in, bankrupt the company to shirk the debt of training / R&D, and hold on to the models in a restructuring. +Enshittification to squeeze maximum profit. This is why they're referred to as vulture capitalists.
Doesn't OpenRouter prove that inference is profitable? Why would random third parties subsidize the service for other random people online? Unless you're saying that only large frontier models are unprofitable, which I still don't think is the case but is harder to prove.
This is one of the reasons why I'm surprised to see so many people jump on board. We're clearly in the "release product for free/cheap to gain customers" portion of the enshittification plan, before the company starts making it completely garbage to extract as much money as possible from the userbase
Having good quality dev tools is non negotiable, and I have a feeling that a lot of people are going to find out the hard way that reliability and it not being owned by profit seeking company is the #1 thing you want in your environment
> but as far as I can see all the AI companies are not making money on inference alone.
This was the missed point on why GPT5 was such an important launch (quality of models and vibes aside). It brought the model sizes (and hence inference cost) to more sustainable numbers. Compared to previous SotA (GPT4 at launch, or o1/3 series), GPT5 is 8x-12x cheaper! I feel that a lot of people never re-calibrated their views on inference.
And there's also another place where you can verify your take on inference - the 3rd party providers that offer "open" models. They have 0 incentive to subsidise prices, because people that use them often don't even know who serves them, so there's 0 brand recognition (say when using models via openrouter).
These 3rd party providers have all converged towards a price-point per billion param models. And you can check those prices, and have an idea on what would be proffitable and at what sizes. Models like dsv3.2 are really really cheap to serve, for what they provide (at least gpt5-mini equivalent I'd say).
So yes, labs could totally become profitable with inference alone. But they don't want that, because there's an argument to be made that the best will "keep it all". I hope, for our sake as consumers that it isn't the case. And so far this year it seems that it's not the case. We've had all 4 big labs one-up eachother several times, and they're keeping eachother honest. And that's good for us. We get frontier level offerings at 10-25$/MTok (Opus, gpt5.2, gemini3pro, grok4), and we get highly capable yet extremely cheap models at 1.5-3$/MTok (gemini3-flash, gpt-minis, grok-fast, etc)
Even if I wrote 10k lines of code a day I will be out of things to implement in day two. Maybe I am working on wrong codebases, but I never felt like I am behind because of not enough lines got writen. It was always finding and forming correct perspective.
Just for fun I checked the codebase stats for platform I am working on and it is 70k lines of code. We are talking about enterprise saas processing payments in billions per year.
Similar story in Europe with white tailed eagles, which are quite similar in size. They were extinct in my area for maybe 60 years and recently returned and even started to hatch.
Universities in eastern bloc were really elite places. Only low single digits percent of people were able to enroll. Also majority of degrees were in STEM, education or medicine as they were deemed useful for the state. To get degree outside of STEM, political background of your family was checked and things like having family member (even say uncle) who emigrated outside of country or having grandparents who owned businesses or farm decades ago will get you discarded. So the smart kids usually have very limited path forward, so STEM it was (if you were lucky)
100 or even 200 will be quite a hard target to hit even considering BOM. Most bare bone optics setup will cost you maybe 100, 130 USD. Unless you find some spare parts or go DVD grating route. The best way is to search for ocean optics parts on eBay. There are multiple "benches", sensors and optics sets available most of the time. Still, it will be more than 200.
Well, lets see how all the economics will play out. LLMs might be really useful, but as far as I can see all the AI companies are not making money on inference alone. We might be hitting plateau in capabilities with money being raised on vision of being this godlike tech that will change the world completely. Sooner or later the costs will have to meet the reality.
reply