Glossary
What is RAG?
A way to make an AI answer from your own documents: it first looks up the relevant pages, then writes the answer from them.
Also written as retrieval augmented generation.
RAG stands for retrieval augmented generation. Instead of relying only on what the model learned in training, the system first searches your own material (policies, price lists, past proposals, manuals) for the passages that matter, and hands them to the model with the question.
Why it matters
It is the difference between a chatbot that improvises and one that answers from your real information, can point to where the answer came from, and stays current when you update a document.
Where you meet it
Most business chatbots and internal assistants that “know your company” are built this way. When you upload files to a project in ChatGPT or Claude, or use NotebookLM, you are using a simple form of RAG.