GEO glossary
Retrieval-augmented generation (RAG)
Retrieval-augmented generation (RAG) is a technique where an AI system retrieves relevant documents at query time and uses them to ground its generated answer. Many answer engines retrieve sources this way to cite current information.
Also known as: RAG
For GEO this is why crawlability and quotable content matter: if a model can retrieve and lift your page, it can cite you; if it can't reach or parse it, it can't.
Related terms
- Grounding — Basing an AI answer on retrieved sources.
- AI citation — A brand or source being named or referenced in an AI answer.
- llms.txt — A published file that guides AI systems to your key content.
See where AI names you today.
Get a free AI visibility audit and find out whether the answer engines name your brand, or a competitor.