Core AI Concepts
Retrieval-augmented generation (RAG)
RAG is an architecture where an AI retrieves relevant documents from a knowledge base and includes them in the prompt — grounding answers in your data, and exposing your data to the prompt.
Retrieval-augmented generation pairs a model with a search step: the system finds relevant passages in a document store, injects them into the prompt as context, and the model answers from them. It is how enterprise assistants answer from your wiki, contracts, or tickets without retraining.
RAG inherits a permissions problem: the assistant retrieves whatever its index can see, so over-broad sharing becomes over-broad answers. It also moves documents into prompts automatically — the payload exists even though no human pasted anything.
Where this shows up
Related terms
See it in your own organization.
Sanitized AI inventories the AI tools in use and redacts sensitive data from prompts before it leaves.