Source grounding
Source grounding ties each claim in an answer to a specific document the AI read.
What it means
Source grounding means the answer is built from retrieved material rather than from the model's memory alone. The tool gathers passages first, then writes from them. A grounded claim can be traced to a page, a table, or a dataset, which makes checking practical.
Grounding is necessary but not enough. A tool can cite a real page that does not say what the sentence claims. Good grounding systems also compare the claim with the cited text. When you review a report, pick three claims at random and open each source. This takes a few minutes and shows you more than any score. A useful habit is to note which step of the process produced each claim you rely on. Treat it as one input among several, not as proof on its own.
Continue this reading path
Cite this entry
https://bestresearchai.com/glossary/source-grounding/