Context
Chen and Koudas report a study of 158,099 AI answers and 468,037 citations in two commercial markets, combining within-website analysis, API search replay and controlled experiments. The study finds that retrieval and semantic question matching explain citation behavior better than several common page-format signals in its tested settings.
What changed
Introduces a large empirical analysis separating query rewriting/retrieval, content properties and model prior knowledge in citation selection.
Why it matters
AI visibility measurement should distinguish retrieval eligibility and generated subqueries from answer citation and brand mentions; formatting-only interventions cannot be assumed to change selection.
Visibility OS interpretation
Visibility OS interpretation: in the studied conditions, query-level retrieval is an important intermediary between content and citation. These results are not a ranking-factor disclosure by any platform.
What remains uncertain
How well do the results replicate across languages, topics and consumer interfaces? How stable are subqueries and brand prior effects as models change?
Evidence claims
In Chen and Koudas's two-market study, ChatGPT API retrieval explained 32% of within-site citation variation compared with 2.8% explained by measured page properties; these are study-specific estimates.
Related historical records
Research traces ChatGPT Web retrieval-to-citation behavior across model settings
That record related to this event.