Glossary · AI in support

What is RAG (retrieval-augmented generation)?

RAG (retrieval-augmented generation) is a way of making an AI model answer from specific documents: for each question it first retrieves the most relevant passages from a knowledge source, then writes the answer from them. Asked "Can I return sale items?", a RAG support agent finds your returns policy and answers from that text.

Also calledRAG · Retrieval augmented generation

By , pen name of the Asks founderUpdated

How it works, step by step

  1. 1Index: your content (help articles, website pages, PDFs) is split into chunks and stored so it can be searched by meaning, not only by keywords.
  2. 2Retrieve: for each customer question, the system finds the chunks most likely to contain the answer.
  3. 3Generate: the model writes a reply using those chunks as its evidence, ideally citing them.
  4. 4Fall back: if nothing relevant turns up, a well-built agent says so or hands off instead of guessing.

Where the term comes from

The name comes from a 2020 research paper by Patrick Lewis and colleagues, "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks", which paired a language model with a searchable index of Wikipedia. It's now the standard design for support bots that answer from a company's own content.

What RAG does and doesn't fix

RAG reduces AI hallucinations because the model answers from your documents rather than from memory, and it lets you change an answer by editing content instead of retraining a model. It doesn't fix bad content: if your knowledge base is wrong or silent, the answer will be too. Retrieval can also pick the wrong passage, which is why testing matters; our guide to chatbot hallucinations covers the safeguards.

In the product

How Asks handles this

Asks works this way: it extracts the text from your website, articles and documents, indexes it, retrieves the relevant chunks for each question and answers from them, showing its sources in web chat. With Shopify connected, product questions are answered live from the store instead. See how it works, or train an AI chatbot on your own data.

7-day free trial · card required · cancel anytime

FAQ

Questions, answered.

01Is RAG the same as fine-tuning?

No. Fine-tuning changes the model itself by training it on examples. RAG leaves the model as it is and gives it relevant documents at question time, which makes answers easier to update and to trace back to a source.

02Does RAG stop AI from making things up?

It reduces it but doesn't eliminate it. The model can still misread a passage or fill a gap. Pair RAG with visible sources, a handoff when nothing relevant is found, and tests before changes go live.

How we made this page

Written by Ray (the pen name of the Asks founder) from how support teams use the term and how Asks implements it, with AI-assisted drafting. Formulas and examples are worked arithmetic, not benchmarks; every outside fact links to its source with the date we checked it, and product facts were checked against Asks on 2026-11-17. How we write and correct pages is in our editorial policy.

Answer customers instantly, hand the rest to your team.

Asks learns your website and help center, answers on chat, email and messaging apps, and hands conversations to your team in one inbox when a person should take over.

7-day free trial · card required · cancel anytime

Chat on WhatsApp