Retrieval-Augmented Generation (RAG)
Retrieval-Augmented Generation, usually shortened to RAG, is a technique that lets an AI look things up before it answers. Instead of relying only on what it memorised during training, the system first searches a chosen knowledge base, such as a company's documents, for relevant information, then uses what it finds to write its reply. This makes answers more accurate, more current, and far less likely to be made up.
A plain language model is a bit like a knowledgeable person answering from memory alone. They know a great deal, but their knowledge is frozen at the moment their training stopped, and when they are unsure they may still answer with total confidence. RAG gives that person a reference library and a good librarian. Before responding, the system fetches the most relevant passages from a trusted collection of documents and reads them, so its answer is grounded in real, checkable material rather than hazy recollection.
In practice, the knowledge base is usually prepared in advance by turning documents into embeddings, which are numerical representations that let the computer find passages by meaning rather than by exact keywords. When you ask a question, the system finds the closest matching passages, hands them to the language model alongside your question, and asks it to answer using that supporting text. The model still writes the reply in fluent prose, but now it has the facts in front of it.
RAG matters enormously for real world AI products. It is how a chatbot can answer questions about your specific policies, a legal tool can cite the right clause, or a support assistant can stay up to date without retraining the whole model. Crucially, it reduces hallucination, the tendency of models to invent plausible sounding but false information, which is one of the biggest barriers to trusting AI in serious settings. Many of the UK companies we cover build their products around some form of RAG.