Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG) is a technique where an AI model fetches relevant documents first, then writes its answer based on them.
Updated · Citable
What Retrieval-augmented generation (RAG) means
A language model on its own answers from what it learned in training, which has a cutoff date and gaps. RAG adds a retrieval step: search an index (the web, or a company's documents), pass the best passages to the model, and have it answer from them, often with citations.
ChatGPT search, Perplexity, Gemini, Claude's web search and Google's AI features all work on this principle. It is why being findable in search indexes now also decides whether you appear in AI answers.