RAG (Retrieval-Augmented Generation)
Retrieval-Augmented Generation (RAG) is a method in which an AI language model retrieves current external sources before composing an answer and builds them into the generated response as a factual basis, instead of relying on trained knowledge alone.
In practice
Google AI Overviews, Perplexity and many company chatbots follow the same basic principle: they retrieve suitable sources before answering and base the answer on them, which makes it more current and more verifiable than a language model on its own could be. For website owners this means that only content which is technically retrievable and recognisable as a self-contained passage can enter the retrieval phase at all. The criteria by which providers then select from it are not published.