Grounding
Definition
Grounding is connecting an AI model's answer to external sources, such as search results, so that claims can be supported and cited.
Why it matters
Grounded answers can show where information came from. Ungrounded answers rely on training knowledge alone and may be out of date.
Example
An assistant that runs a web search and shows numbered citations is grounding its answer.
Related terms
Concepts
Retrieval augmented generation (RAG)
Retrieval augmented generation is a technique in which a system retrieves relevant documents and gives them to a language model to use when writing an answer.
Sources
Citation rate
Citation rate is the share of eligible AI answers that show a link to a brand's own pages as a source.
Concepts
Large language model (LLM)
A large language model is an AI system trained on very large amounts of text to predict and generate language.
What is Grounding?
Grounding is connecting an AI model's answer to external sources, such as search results, so that claims can be supported and cited.
Why does grounding matter?
Grounded answers can show where information came from. Ungrounded answers rely on training knowledge alone and may be out of date.
See these terms measured on your brand.
Start with a first audit. We show where you appear, who is named instead, and what to fix first.