glossary term · cross business
Retrieval-augmented generation
Learn what retrieval-augmented generation means through a plain-language definition, a synthetic business example, practical limits, and links to workflow guides.
Definition
Retrieval-augmented generation, often shortened to RAG, combines finding relevant material in an approved source collection with generating an answer using that material. The retrieval step selects evidence for the current request. It can improve access to relevant information, but it does not guarantee the selected material is current, complete, authorized, or correctly represented in the answer.
Owner example
Synthetic example: A support request triggers a search of approved policy entries. The system returns the matching guide-access passage with its source ID, then prepares a draft for review. The reviewer checks both the passage and the claim.
Non-example and limitation
Supplying the same short FAQ in every request is not necessarily a retrieval system: nothing may be searched or selected. A retrieved passage can still be stale, irrelevant, or contain untrusted instructions.
Related terms
The context window limits material available in a request. Hallucination can still occur when a source has been retrieved.
Deeper guide and practical link
Read support knowledge preparation for the fuller explanation. Use the supplied-FAQ support workflow to apply the distinction to a bounded task.
Sources and scope
This is an original editorial definition with a synthetic example, not a report of a tested implementation or business result. Suggest a correction.
Change notes
Foundation draft prepared October 20, 2026. No public publication or successful live-tool test is claimed.
Next learning step: Prepare your support knowledge before connecting AI