The short answer
Retrieval-augmented generation (RAG) is an AI architecture in which the system first searches a knowledge source — such as your policies, manuals or product data — and then generates an answer based on the passages it found, usually with citations.
Why RAG matters
An LLM on its own only knows what it learned in training. RAG gives it access to current, private and specific information without retraining, and makes answers verifiable because every answer can point to its source.
How a RAG system is built
- Documents are split into passages and converted into embeddings stored in a vector database.
- A user's question is converted the same way and used to retrieve the most relevant passages.
- The model receives the question plus those passages and is instructed to answer only from them.
- The answer is returned with citations, and access controls filter what each user may see.
Where RAG works well
Internal knowledge assistants, customer support over help-centre content, policy and compliance Q&A, sales enablement and document-heavy workflows in legal, real estate, education and healthcare administration.