RAG

RAG bolts a search step onto a language model: before answering, the system retrieves relevant documents from your own data and hands them to the model as context. It grounds answers in real sources, reduces hallucination, and keeps proprietary data out of model training. The security catch is that the retrieval store becomes a new attack surface: whoever can write to it can influence answers, and the access controls on the documents must survive the trip into the prompt.