Skip to content

Glossary · Updated Sep 18, 2026

Retrieval-augmented generation (RAG)

Retrieval-augmented generation is a technique where an AI system first finds relevant passages in approved sources, then uses them to write an answer with references.

RAG is how most company knowledge assistants work. Instead of relying on what a model learned in training, it answers from your SharePoint sites, policies, contracts or product documentation, and it can point the user to the source so the answer can be checked.

For example, an employee could ask about the travel expense policy and get a short answer linked to the current policy document. The limits are practical. Answers are only as good as the documents retrieved, so outdated, duplicated or conflicting files produce poor answers, and the system must respect the same access permissions as the original sources. RAG reduces hallucination but does not remove it.

See how an idea becomes a working solution.

Explore how we scope, build and hand over AI projects, with clear decisions at each stage.