# Grounding

> Grounding connects an AI response to supporting information, such as retrieved documents or a trusted data source.

*Source: [https://contentcucumber.com/aeo-glossary/grounding/](https://contentcucumber.com/aeo-glossary/grounding/)*

*Topic: AI search basics. Also called: Grounded answer*

---

## Why it matters

A response can use current external material, but readers still need to check that the material supports the claim.

## Example (illustrative, not a client result)

An assistant consults a current returns policy before answering a question about the return window.

## What to check

Open the cited source, check its date and scope, and compare the answer with the original wording.

## What this does not tell you

Grounding can improve factual support without making every statement correct, since a retrieved source can itself be wrong.

## Sources

- [Google Cloud: generative AI glossary](https://docs.cloud.google.com/docs/ai-ml/generative-ai/glossary) (Platform documentation)

*Evidence type: Documented concept. Sources checked on 2026-11-19*

## Related terms

- [RAG](https://llms.contentcucumber.com/aeo-glossary/rag.md)
- [Hallucination](https://llms.contentcucumber.com/aeo-glossary/hallucination.md)
