Definition
RAG (Retrieval-Augmented Generation) is a technique that enhances LLM responses by first retrieving relevant information from external knowledge bases, then using that context to generate more accurate and up-to-date answers. It combines the strengths of information retrieval with text generation.
Example
An AI agent answering questions about your company's policies first searches the internal wiki (retrieval), then generates a response grounded in those specific documents rather than relying solely on its training data.
Why it matters
RAG solves the two biggest problems with LLMs: hallucination (making things up) and stale knowledge (not knowing current information). Gartner predicts that by 2028, enterprise search software will predominantly use RAG to ground responses in real data.