[grumbel]

It's weird, I always expected AI categorization and search to arrive long before generative content. But despite a lot of research going into that direction, it never really arrived in the consumer space. Search today feels not fundamentally different than search 20 years ago and if I want to know about a movie I still use IMDb like it's the 90s.

The already enormous mountain of content out there keeps growing, fuel by AI now, yet ways to explore that mountain keep staying the same or even diminishing, as the clutter in the results just keeps increasing.

[nvarsj]

I think the issue is LLMs don't scale well to the level needed for search. They also have obvious latency. So generative usage seems to be the best fit at the moment.

[grumbel]

But you could feed the text through the LLM, generate a summary and a list of keywords, characters and locations and search through that the regular way. And as far as I can tell, that doesn't really exist yet.

IMDb has a keyword feature[1], but that's still completely human-curated and quite useless due to that, as it just doesn't have the necessary depths and completeness to discover interesting new content.

[1] https://www.imdb.com/search/keyword/