AI Support & Agents

Retrieval-Augmented Generation (RAG)

Definition

Retrieval-augmented generation (RAG) is a technique where an AI model retrieves relevant information from your own knowledge sources at query time and uses it to ground its answer, instead of relying only on what it memorized during training.

Also known as: RAGgrounded generationretrieval augmentation

How it works

When a question comes in, the system searches a knowledge base — help-center articles, docs, past tickets — for the most relevant passages, then passes those passages to the language model along with the question. The model writes its answer using that retrieved context.

This is why a RAG-based support agent can answer questions about your product, pricing, and policies even though the underlying model was never trained on them.

Why it matters for support

RAG is the main defense against hallucination in customer support. Grounding answers in your real content keeps them accurate and current — but the quality of the answer is only ever as good as the sources you connect. Clean, non-contradictory documentation produces confident, correct answers; stale or conflicting content produces confident wrong ones.

Frequently asked

Does RAG stop AI hallucinations?

It sharply reduces them by grounding answers in retrieved source content, but it doesn't eliminate them — answer quality still depends on how clean and current your knowledge sources are.

What does RAG stand for?

Retrieval-augmented generation — retrieving relevant source material and using it to augment (ground) the model's generated answer.

Connect your knowledge sources

Related terms

Agentic AI

Agentic AI is AI that doesn't just answer questions but takes actions — it plans multi-step tasks, calls tools and APIs, and works toward a goal end to end with little or no human intervention..

AI Agent

An AI agent is a software system that uses a language model to understand a request, decide what to do, and take actions — calling tools, retrieving data, and completing tasks — rather than just returning a single scripted answer..

AI Hallucination

An AI hallucination is when a language model generates a response that is fluent and confident but factually wrong or fabricated — inventing details, policies, or sources that don't exist..

Knowledge Base

A knowledge base is a structured, searchable library of articles — how-tos, FAQs, troubleshooting guides, and policies — that lets customers or agents find answers without contacting support directly..

Put these ideas to work

Macha is an AI agent layer that sits on top of the help desk you already run — Zendesk, Freshdesk, Front, Intercom, or Gorgias.

Start Trial