Glossary
What Is RAG (Retrieval-Augmented Generation)?
RAG (Retrieval-Augmented Generation) is a technique where an AI system first retrieves relevant information from a specific set of documents — your own docs, FAQs, or other content — and then uses that retrieved information to generate its answer, instead of relying only on its general training.
Language models are trained on a broad, general dataset up to a fixed point in time — they don’t automatically know about your specific product, policies, or documentation. RAG closes that gap: when a question comes in, the system searches your actual content for the most relevant pieces, and gives those to the model as context before it writes an answer.
This matters for accuracy. Without RAG, an AI might confidently guess at an answer based on general patterns. With RAG, it grounds its answer in the actual source material you provided, and can even point to where that information came from.
This is the mechanism behind letting a chatbot answer from your documentation, product pages, or even a video walkthrough: the content gets processed and indexed, and retrieved at the moment a relevant question comes in. Basium’s ingestion tools (for text and video content) work on this principle.