> Also 20 billions of revenues, not profits, is orders of magnitude too low compared to their expenses.
Nah, it's just one order of magnitude...
Also, they expect revenue to grow exponentially so it's 20 billions annualized by the end of the year. Last time I saw somebody talk about it, it was about half of it, and trending down.
Anyway, if they manage to take ~20% of the ads revenue from Google, they will be able to cover ongoing depreciation! That's the amount of money they need.
The problem is that inference is a whole different ballgame in terms of costs compared to a traditional SaaS model where each extra customer adds near zero in cost.
They may make it work but OpenAI is more akin to a traditional high revenue low profit business like for example a grocery store.
Thats why we are seeing the explosion of extra tools to try lock in business for higher value use cases and not fight on the margin.
Nah, it's just one order of magnitude...
Also, they expect revenue to grow exponentially so it's 20 billions annualized by the end of the year. Last time I saw somebody talk about it, it was about half of it, and trending down.
Anyway, if they manage to take ~20% of the ads revenue from Google, they will be able to cover ongoing depreciation! That's the amount of money they need.