Retrieval
Also called information retrieval.
Retrieval is finding and returning information relevant to a request from stored or connected material, so an AI app can use it as contextContext is the information available to an AI model for a particular response, including instructions, conversation, supplied material, and tool results..
Example
For a question about venue parking, the app selects the parking paragraph from a handbook instead of passing the whole handbook into every answer.
- A question“How do I cancel a booking?”
- Search the sourcesHandbook, files, earlier chats.
- Relevant passagesThe cancellation section, not the whole handbook.
- Write the answerThe model uses those passages and links to them.
Why it matters
An app can retrieve passages, records, images, or earlier conversations. The search method affects what is found and what useful material is missed.
Retrieval only finds material. Writing the answer is a separate step. When an app does both on purpose, the pattern is called retrieval-augmented generationRetrieval-augmented generation, or RAG, finds relevant source material and gives it to an AI model before the model writes its answer..
Common confusion
A missing search result does not prove the source lacks the answer. Keyword search, similarity search, and combined search can return different results.
Sources
Updated October 1, 2026.