Context
Benjamin Tannenbaum reports a June 6, 2026 observational audit of 15 fixed commercial prompts across ChatGPT, Microsoft Copilot, Google and Perplexity. The study reports 589 citation observations representing 528 distinct URLs and 356 domains; mean pairwise same-prompt exact-URL Jaccard overlap was 0.0079, and 84.9% of compared engine pairs shared no cited URL.
What changed
A newly discovered original preprint measures engine-specific retrieval/citation exposure and explicitly separates page-fit scores from live exposure and conditional citation selection.
Why it matters
Cross-platform AI visibility dashboards need to distinguish page/query fit, observed source exposure, conditional citation selection and final citation presence, rather than assuming a platform-neutral page score predicts mentions.
Visibility OS interpretation
Visibility OS interpretation: a single page-quality or engine-agnostic GEO score is not an adequate substitute for observed, platform-specific citation tracking in this limited audit. This is a measurement insight, not a confirmed search-ranking mechanism.
What remains uncertain
Do these overlap and turnover estimates persist across larger, representative prompt samples, languages, later model versions, and repeated collection dates? The 15-prompt sample and June 2026 snapshot restrict generalization; full manuscript was not directly fetched in this cycle.