Retrieval-augmented generation: grounding AI answers in your documents

Staff spend a lot of time searching for answers in manuals, policies and past reports. Retrieval-augmented generation (RAG) lets an AI assistant answer from those documents.

How it works

1. Documents are split into passages and indexed for search.
2. When a user asks a question, the most relevant passages are retrieved.
3. The language model writes an answer using those passages and cites them.

Why it matters

Answers are based on your current documents, not the model's general knowledge, and users can check the sources.

Keep content current

The assistant is only as good as the documents behind it. Remove outdated versions and assign owners for key documents.

Respect permissions

Users should only get answers from documents they are allowed to see.

Evaluate

Build a set of real questions with known answers and test regularly, especially after changes.

See our AI and knowledge systems work.