They're useful for embeddings, which let you turn articles (and images and other content) into a huge array of floating point numbers that capture the semantics of the content. Then you can use a vector database to find similar items to each other - or similar items to the user's own search query.

I wrote a big tutorial about embeddings a couple of years ago which still holds up today: https://simonwillison.net/2023/Oct/23/embeddings/

finding similar things quickly, where the "shape" of a thing can be defined by a vector (like embeddings for instance). this can be used in lots of machine learning applications

I figured it would be something like this. And vectors as rows in a regular table would be too slow then?

sqlite does not have native support for a vector-like column type. Extensions like this and sqlite-vec build on the BLOB column type, and provide additional functions to let you efficiently search and manipulate this vector data.

Everyone is saying AI, but also things like image similarity search.