Openai is really non profit after all
(midwest.social)
TechnologyWith a ~40 billion usd hole (netloss) Openai keeps it´s word by staying a nonprofit company 🤣
Don't you worry about profits, let ME worry about blank!
I find it interesting that according to these numbers, if they entirely stopped R&D and marketing, they would just about break even.
I'm noticing the way that last year's expenses get covered by this year's revenue. Maybe they'll try for some deal around that, and expect the bubble to last forever?
If stopping R&D and marketing had no impact whatsoever on their future revenue.
Fat chance.
Oh definitely. I think it's anthropic who have stated in multiple interviews that they break even on most of their models, it's just that they keep spending exponentially more to train the next model. They and openai seem to be stuck in an arms race where switching to purely serving existing models to their existing clients just won't work. I do wonder how accurate that assessment is on their part.
There's also DeepSeek eating revenue that would go for those more "mature" models.
I think it's anthropic who have stated in multiple interviews that they break even on most of their models
Words are cheap pre IPO. I don't believe that for a second.
They're anti-profit lol
Those techies might be able to create widgets but they're sure not competent businesspersons.
Ed's article where he released those numbers: https://www.wheresyoured.at/exclusive-openai-financials/
They'll be profitable when the training stops/slows down. But how much revenue they can generate is another question. This might be the reason behind the news of US gov thinking about banning Deepsek and other Chinese models. Open weights models kill the API billing bussiness model. Margins in that scenario probably wouldn't cover the massive investment.
How are they supposed to “slow training down” without falling behind their competitors?
Competitors have the same problem. We'll see what happens once subsidies run out.
So the choice is between unprofitability and models being so out of date they become useless? Great.
https://isaiprofitable.com/
That page makes it look even better than it is. It's almost halfway to profitable, according to that. About 2/3 of that revenue is from Nvidia though, which is just selling the shovels for the gold rush speculation. It's not "AI profit" as if AI is producing something. It's just sales caused by people hoping it'll produce something in the future.
Idk the source of that data but OMG look at anthropic, 6.5b revenue but recently "evaluated" ~1 trillion USD. Is that for real?
Yeah, all speculative.
Ahahahahaha Facebook has dumped roughly the same amount of money into AI as Microsoft, but makes 1/10th the revenue. Pure gold
OpenAI is the definition of the difference between "Not-For-Profit" and "For not-profit".
Which column is infrastructure under? Things like power, DC rent, etc. Also what about depreciation assuming you try to shove all that shit (well the hardware anyway) into CapEx?
Which column is infrastructure under?
I don't think they have any of that. They rent computation capacity, which should partially be under cost of revenue and partially under R&D.
This misapprehends the business models of techno feudalism: you operate at a loss, running on VC money which is fairly endless. You offer a freeish service to grow your brand. Then when you list on the stock exchange you live off debts loaned to you by the bank on the basis that your company is worth millions or billions.
I have no doubt ChatGPT is, and that Sam Altman lives with no taxable income to speak of. He's just accumulating bank loans and paying them back with other bank loans secured by OpenAIs growing stock price and company valuation.
Money isn't real to the elite classes. Actually I believe Stalin and Churchill once stopped at a Cafe sometime after WW2, they realised after their meal no one travelling with them had money on them to pay. Because they didn't need money. It's accepted when you're at that level. Welfare Capitalism has existed for a very very long time.
As real as any valuation. Which is to say, what investors are willing to pay for a piece of a company may not always have the most firm association with the reality of a company's current state. And, the market can stay irrational much longer than you expect.
🤣
