3 ms·
> Author: A paying Max 20x customer who's done subsidizing Anthropic's infrastructure problems Shouldn't that be the other way around? Isn't every LLM provider
by StrLght 8mo ago
> Author: A paying Max 20x customer who's done subsidizing Anthropic's infrastructure problems
Shouldn't that be the other way around? Isn't every LLM provider losing money? Especially on premium subscriptions.
- smashah 8mo agoIs that even publicly knowable? In sure they're not doing charity with their subscription despite our perceptions.
- g947o 8mo agohttps://x.com/sama/status/1876104315296968813?lang=en https://x.com/sama/status/1876104315296968813?lang=en
- bryanlarsen 8mo agoI don't think that's clear. They're certainly burning money, but that's mostly R&D: salaries and training compute. Once you remove those, it's unclear whether the AI companies would be losing money on just inference.
- bombolo 8mo ago[dead]
- dns_snek 8mo agoYou can't "remove" costs willy nilly to make a company look profitable. Running an independent ISP is an extremely lucrative business if you gloss over the capex requirements of installing and maintaining the infrastructure.
- bryanlarsen 8mo agoYou don't remove them willy-nilly, but evaluating companies based on their operating costs is standard accounting practice. R&D is not part of operating costs. Capex for inference is in, capex for training is out for that analysis.
- dns_snek 8mo agoHow much inference are they going to sell a year from now if they stop R&D? Any comparison excluding those costs is either misguided or fraudulent.
- bryanlarsen 8mo agoIt's one of the key metrics of a business. These are growth businesses, everybody assumes they're going to have a lot more customers in the future than they have today. If they've got X times as many customers next year than they have right now, training costs per customer gets divided by X, but inference costs per customer stay roughly constant. So for businesses where they expect X to be large, they care about operating costs a heck of a lot more than R&D.
- energy123 8mo agoThere was some credible analysis that I don't have the link to which estimated 50% gross margins for OpenAI, largely eaten up by operational expenses. So not awful unit economics, but not good either. Assuming that's even true, the big asterisk is uncertainty around efficiency gains in the future. The intelligence divided by cost ratio is changing very quickly. It is hard to make confident predictions more than 3 months out.