What you are missing is that it turns out the gold isn’t actually gold. It’s bronze.
So earliest, the shovelers were willing to spend thousands of dollars for a single shovel because they were expecting to get much more valuable gold out the other end.
But now that it’s only bronze, they can’t spend that much money on their tools anymore to make their venture profitable. A lot of shovelers are gonna drop out of the race. And the ones that remain will not be willing to spend as much.
The fact that there isn’t that much money to be made in AI anymore means that whatever percentage of money would have gone to NVIDIA from the total money to be made in AI will now shrink dramatically.
Perhaps they mean there's less wealth to be extracted from the closed-source training side of the equation, which requires huge capital investment, and promises even bigger returns by gatekeeping the technology.
Many discussed aspects are disconnected. Cost of training, cost of hardware(and margin there), cost of operation, possible use cases, and then finally demand.
Cheaper training still expect there is some use case for those trained models. There might or might not be. It can very well be that cost of training did not really limit the number of usable models.
So earliest, the shovelers were willing to spend thousands of dollars for a single shovel because they were expecting to get much more valuable gold out the other end.
But now that it’s only bronze, they can’t spend that much money on their tools anymore to make their venture profitable. A lot of shovelers are gonna drop out of the race. And the ones that remain will not be willing to spend as much.
The fact that there isn’t that much money to be made in AI anymore means that whatever percentage of money would have gone to NVIDIA from the total money to be made in AI will now shrink dramatically.