alt Hacker NewsI think that the main flaw in the reasoning is assuming that cost of token will stay the same over the years.
Chances are that token prices will go down, but chances also are that the AI bubble pops and all of a sudden all these companies will either have to make a buck out of the inference or go bankrupt.
Getting your own hardware just grants you stable pricing.
if all those companies go bankrupt, you could also buy their hardware on the cheap.
its almost guaranteed imo that as the model quality evens out, inference cost will drop towards 0 at insane speeds, given how well llama works as an asic.