RAG (Retrieval-Augmented Generation) is the technique of giving an AI model access to an external knowledge base at answer time. Instead of relying only on what the model "memorized" during training, the system retrieves the most relevant snippets from your documents and injects them into the prompt, so the answer is grounded in real, up-to-date data.
Why it matters
Language models hallucinate: they invent facts confidently. RAG reduces that risk by anchoring the answer in concrete sources, and it lets the AI answer about information it never saw in training, such as your company's manuals, contracts or support history, without retraining the model.
In practice
A typical RAG flow indexes your documents in a vector database, turns the user's question into a similarity search, retrieves the closest snippets and hands them to the model along with the question. It is the backbone of most enterprise assistants and AI-powered internal search. RAG quality depends more on the data and retrieval than on the model itself.