In my mind the most obvious scenario where AI gets used a lot but AI companies (exlc. NVIDIA / chip makers) aren't worth much is the hardware becomes capable of running the current state-of-the-art models locally, cheaply, and those models are basically "good enough" for 90% of tasks.

Sounds like the dot com bubble to me. many companies went bankrupt but the technology marched on.

P1 is probably necessary but not sufficient for P2, which would show up as a correlation of a sorts. Maybe I should've specified even further, if the tech works (P1 and P1a) then yeah somebody is going to make a lot of money but not necessarily OpenAI/Anthropic. They could go the way of AOL/Time Warner (probably not but again who knows). Microsoft and Amazon survived the dotcom bubble and thrived, Meta came after it.

Currently they don’t seem to be. At least OpenAI is haemorrhaging money left and right. Thing is, the compute for all this is far more expensive than anyone is willing or able to pay.

The most successful use of AI has almost certainly been advertising, which companies were already doing before the recent AI hype.

If by AI we mean LLMs, the question looks a bit different.