Definition
RAG (retrieval-augmented generation)
A technique connecting an AI model to company documents: the AI retrieves relevant passages before answering, with cited sources.
RAG (Retrieval-Augmented Generation) solves the main limitation of LLMs: they don't know your internal data. The principle: your documents (contracts, procedures, wikis, emails) are indexed in a vector database; for each question, the system retrieves the most relevant passages and feeds them to the model, which answers citing its sources.
RAG sharply reduces hallucinations and makes AI genuinely useful at work. Sovereignty caveat: the index contains a condensed copy of your most sensitive documents — it must be hosted inside your perimeter, not in a third party's cloud.
Further reading
Sovereign AI, in practice
Every AI model, deployed on your infrastructure, with your documents connected.