Retrieval-Augmented Generation (RAG)
Retrieval-augmented generation (RAG) is an approach where an AI first looks up relevant content from a specific source — your site, your documents, a database — then uses what it finds to write its answer. Instead of relying only on what the model absorbed in training, it grounds the response in material you actually control.
On a web project this shows up two ways.
- It's how an on-site AI assistant can answer using your real content rather than guessing.
- It's part of why structured, machine-readable content matters for being quoted accurately by an outside answer engine.
Either way, the retrieval is only as good as what it draws from, so a clean content model and solid schema pay off here.
