RAG (Retrieval-Augmented Generation)

The technique AI search tools use to look up real, current information from the web or a database and ground their generated answer in it, instead of answering purely from training memory. Every major AI answer engine runs some version of retrieve, rerank, then generate, which is why content structure and freshness directly affect whether a page gets pulled into an answer.