Data & retrieval
RAG
Retrieval-Augmented Generation — a technique that gives an AI model access to external documents before it answers, so it can cite real, up-to-date sources.
01 ——
In plain English
RAG (Retrieval-Augmented Generation) is a pattern that improves AI accuracy by fetching relevant documents and injecting them into the model's context before it generates a response.
How it works:
- User asks a question
- The system searches a document store (your docs, a database, the web) for relevant content
- That content is added to the prompt sent to the LLM
- The LLM answers using both its training knowledge and the retrieved content
Why use RAG instead of fine-tuning?
- No retraining required — update your documents, not the model
- The model can cite sources, reducing hallucination
- Works with private or frequently-changing data
RAG is the most common way to build AI tools that "know" your company's information.
02 ——
Tools that use RAG
03 ——
Related terms
LLM
Large Language Model — the type of AI behind tools like ChatGPT and Claude, trained to understand and generate text.
Embeddings
A way of converting text (or images) into lists of numbers so an AI can measure how similar two pieces of content are.
Vector Database
A database optimised for storing and searching embeddings (numerical representations of text or images) by similarity.
Hallucination
When an AI confidently states something that is factually wrong or completely made up.
Semantic Search
Search that finds results by meaning rather than exact keyword matches — so "car" finds results about "automobile" too.
Back to glossaryLast reviewed May 2026