Posted by tolugenius 14 hours ago
Imagine if running fable costs you what it actually costs to run fable. A lot of vibe coders (and just proper software engineers) are gonna be very sad if that comes to pass.
But that reminds me of the story of the union rep telling Ford: “good luck getting your machines to buy your cars”
(Fyi Ford took note and started paying his workers enough that they’d buy his cars)
https://en.wikipedia.org/wiki/Islamic_banking_and_finance
Imagine a scenario where the AI bubble bursts and AI companies and neoclouds go bankrupt en masse, and then a huge rebound occurs when AI has a delayed takeoff. Nvidia ends up with a massive amount of compute on its hands from its backstop deals, and it also owns assets from failed companies when profits start to grow. New startups running using Hugging take the place of OpenAI and Anthropic when their compute assets are divided between survivors like Nvidia, Microsoft, Alphabet, Meta.
If/when there is an AI crash, any number of small startups can buy compute for the price of electricity without anyone wanting to buy them. That is when the real innovation happens. The top of the hype cycle is usually more about getting rich quick and buying and shutting down competition.
Nvidia booked $496 million in interest income in Q2 alone [1].
[1] https://www.sec.gov/Archives/edgar/data/1045810/000104581026... page 15
And then when the kings ran out of money, they sacked the Knights and whoopsies, looks like the knights had no real money as they spent it as quickly as they could maintaining their financial network... Sound familiar?
If demand vanishes for the 3 million cards Amazon just bought, then something will be done with them. The AI market may end up in a bizarre jepson's paradox of rotation between inference use cases and model training.