If I had to guess, probably Web-style.

LLMs are actually useful, people are willing to pay for access to them, and they do genuinely enable things that were unrealistic or impossible before. The advances in image, video, and sound models since 2020 are also striking but likely won't be as transformative as LLMs.

That being said, I don't think it's unlikely that we'll see a plateauing of progress followed by a strong crash/correction in the market a-la dot com. The Allbirds situation absolutely has echoes of pets.com.

I also feel that commodification is coming for models, training/inference hardware, and software (e.g. CUDA), as it has for nearly everything else useful in tech. So I expect valuations driven by unique advantages here to be eroded over time (Think Sun and SCO after Linux on cheap x86 servers became the norm).

I think it will be like the 90's collapse but bigger. It must be noted that the financial markets have been awash in cash for some time since companies have been making huge profits since the 90's. So the money that will be lost will mostly be that of rich investors and large companies who were sitting on a bunch of capital they didn't know what to do with.