AI Glossary
What is Retrieval-Augmented Generation (RAG)?
A technique where an AI system first finds relevant information from a set of documents or a database, then uses it to write its answer, so responses are based on your actual content rather than only on what the model learned in training.
A plain language model answers from what it learned during training. With RAG, the system first retrieves relevant passages, for example from your company's policies, product manuals or help articles, and then generates an answer using them, often with references.
Why it helps:
- Answers can use up-to-date and company-specific information.
- Fewer invented answers, because the model works from real text.
- Answers can cite where the information came from.
Where you see it: AI help bots on websites that answer from a company's documentation, tools like NotebookLM that answer only from the sources you add, and chat assistants that search the web before answering.
You can use the same idea in any chat assistant: paste the relevant document and say "Based only on this text…". See our NotebookLM guide.