All of them sucked, though. Turns out, running regression on some features picked as much to help the user as to satisfy business objectives, doesn't lead to a system that can capture one's preferences well.

LLMs have the benefit of understanding (or some values of understanding) how people feel about things in general, in a global sense. They may not have a database of people's choices, but they have a "database" of connotations for every word, how ideas and emotions relate, how interests connect, etc. Instead of relying on a relatively tiny historical record of choices in few, specific (and ill-defined) categories, they can just place user's history in their 10 000 dimensional latent space and use that as a direction to explore, effectively guessing whatever the user's actual preferences are likely to be, without being able to name them or fit into explicit categories.

yeah I mentioned in another comment I've never found recommendation systems to work very well for me. I've gone through many of them and the reason I decided to start using LLMs was because I was out of options...and after I tried it I ended up much preferring the recommendations given.

You can also specify more granular in human words what you're looking for which is a big bonus for me personally.