Retrieval-Augmented Generation (RAG)
Retrieval-Augmented Generation (RAG) is an approach where an AI first retrieves relevant passages from a specific knowledge source, then uses only those passages to write its answer — grounding the response in real information rather than the model’s memory.
How it works
A plain language model answers from patterns it learned during training, which means it can confidently state things that are outdated or simply wrong. RAG fixes this by giving the model a fresh, trusted source to work from at the moment of answering.
When a question comes in, the system searches a knowledge base for the passages most relevant to it, hands those passages to the model, and instructs it to answer using only that material. The model becomes a careful summarizer of real content rather than a source of invented facts.
Because the answer is tied to retrieved passages, you can also show exactly which sources were used — making the response verifiable.
How SiteMind uses it
SiteMind is built entirely on RAG. It retrieves passages from your own website and documents, answers strictly from them, and cites the source pages — which is why it can refuse to guess when the answer simply isn’t in your content.
Related terms
Embeddings
Embeddings turn text into lists of numbers that capture its meaning, so a computer can tell which pieces of content are about the same thing.
Semantic search
Semantic search finds content by meaning rather than exact keywords, so it can match a question to the right answer even when the words differ.
AI hallucination
An AI hallucination is when a model states something false or made-up as if it were true — a serious risk for any customer-facing chatbot.
Put this to work on your site
Free 3-day trial, no card required.