1. The AI market can support many players.
What is interesting about the AI/LLM boom is how it can support so many entrants and big players–like Cursor AI, Open AI, Cline, Gemini, Claude, Grok, CoPilot etc.–and each carving a niche and a high valuation. This is in contrast to social networking, in which Facebook was the overwhelming dominant player, followed a distant second by LinkedIn (Instagram in 2012 was bought out by Facebook). Or Yahoo vs. Google.
But now it’s like a dozen companies, and each worth a lot and not as interchangeable, compared to as seen with search engines or social networking. Facebook was the clear and dominant winner and superior platform, and there was little reason to use an alternative, except maybe LinkedIn for job searches. Same for Google–it made little sense to use Yahoo or Bing. The closest competitor to Google, Duck Duck Go, is still a tiny niche.
This portends favorably to AI NOT being a bubble or a fad, because it indicates that AI is seeing integration with many facets of society and the economy. There is so much demand for AI that ‘the market’ is able to accomodate all these companies, instead of the winner-take-all paradigm otherwise seen with past bubbles or fads.
Compare this to an obvious bubble, such as Beanie Babies, in which there was only a single supplier, that being Ty Warner, and no integration with the overall economy. Even at the peak of the Beanie Baby bubble in 2000, Beanie Babies could have vanished overnight and there would be no secondary effects on the U.S. economy. By comparison, if Eliezer Yudkowsky gets his wish and LLMs vanish, companies (that use AI for enterprise purposes) and end-users (such as inexpensive Chat GPT subscriptions), would notice immediately. It would be a major shock felt on the stock market and probably the economy too.
2. No use for cryptocurrencies.
It would seem befitting for cryptocurrencies to have a role in AI given the chronological overlap and the large overlap between users and fans of AI and crypto, but technologically, this has not been the case. There has been no ‘spillover effect’ of AI into crypto. The price of Ethereum is unchanged since November 2021 despite AI valuations surging. As far as the market and speculators is concerned, these are totally separate technologies and cycles. I think however crypto is much more of a bubble compared to AI, for the reasons given in the above section. AI has seen much more economic integration compared to crypto, which is still mostly for speculators and gamblers, whereas AI is used for serious work.