Groundedness
Whether a model's output is actually supported by the context it was given, rather than invented.
Groundedness measures the relationship between a model's response and its source material — provided documents, retrieved passages, or conversation history — asking whether every claim in the output can be traced back to that material. An ungrounded response might still be factually true, but if it isn't supported by the given context, it's still a groundedness failure, since the model should have said "I don't know" or asked for more information instead.
This is closely related to but broader than fact-checking: fact-checking verifies specific claims against real-world truth, while groundedness verifies claims against the specific context the model was actually given, which matters enormously for RAG systems where the whole point is to answer from retrieved documents rather than parametric memory.
Ungrounded responses are one of the most common root causes flagged in hallucination evaluation.
What this means for trainers
Always check the response against the provided source text line by line rather than your own general knowledge — a groundedness failure can happen even when the model's claim happens to be true.
Related terms
Put this into practice
Browse open AI training roles from Alignerr, Mercor, Outlier, and more.
Browse AI training jobs