Retrieval-Augmented Generation (RAG)
Retrieval-augmented generation fetches the most relevant source passages and hands them to the AI before it answers, so replies stay factual.
Definition
Retrieval-augmented generation, or RAG, is a technique where the system first searches a body of sources — your uploaded documents, a knowledge base, or the live web — pulls out the passages most relevant to your question, and adds them to the prompt before the model answers. Imagine asking about a 40-page insurance policy: instead of guessing, the tool finds the two paragraphs that mention your exact coverage and answers from those.
Why it matters
RAG is how an assistant can answer questions about your private files or this week's news, neither of which its training ever saw. It is the engine behind grounding: fetch the right passages, then answer strictly from them, ideally with citations you can click. In AskClear's Facts mode, your documents and web results are retrieved this way, so an answer about your lease points to the actual clause rather than a vague memory.