Retrieval-Augmented Generation (RAG) is a method where an AI model fetches relevant documents at answer time and grounds its response in them, rather than relying only on what it memorized during training. It is the mechanism behind many cited AI answers.
In plain terms
A RAG system retrieves documents related to a query and supplies selected context to a generative model before it produces an answer.
Some answer products expose source links or grounding metadata, but a linked answer alone does not reveal every implementation detail or prove that one specific RAG architecture was used.
Accessible, relevant, and clearly structured content can be easier to retrieve and interpret; retrieval still depends on the provider's index, query, ranking, and controls.
Why it matters
RAG explains how a system can use information that was not present in a model's original training data.
It also separates two questions: whether a document was retrieved and whether the generated answer represented it accurately.