What is Ground-truth surface?

A ground-truth surface is the single canonical page a brand designates as the loud, unambiguous, machine-readable statement of the facts an AI engine keeps getting wrong — current pricing, current features, founding facts, and explicit disambiguation (‘we are not [the other brand]’) where conflation is happening.

Usually hosted on the entity home, it is the retrieval target that a fact-correction effort points every engine at: when an engine tries to ground a brand fact, the ground-truth surface is what it should find first and prefer over its stale training memory. It works by pairing plainly-stated current facts with the right schema — Product and Offer for pricing, Organization for founding facts, FAQPage for the exact question-and-answer form the brand-fact query set uses — so the correct fact is unambiguous to a parser, not just a human reader. The ground-truth surface is the textual analog of grounding a generative model on a real reference: rather than correcting a wrong fact after the model states it, it makes the current fact so retrievable and structured that the engine adopts it. It is the primary lever for retrieval-time and stale-fact hallucinations, and a necessary (if slower-acting) support for training-baked ones.