Context
An original longitudinal audit issued 55,393 trending queries across 19 categories and examined when Google AI Overviews appeared, which domains they cited, and whether cited text supported generated claims.
What changed
The study measured 13.7% AI Overview activation overall and 64.7% for question-form queries; 29.8% of cited domains were absent from co-displayed first-page results, and 11.0% of 98,020 assessed atomic claims were contradicted or not addressed by retrieved cited source text.
Why it matters
Conventional first-page rankings, AI Overview triggering, citation inclusion, and claim fidelity are different visibility and quality measurements; one should not be substituted for another.
Visibility OS interpretation
Visibility OS interpretation: independent, separate AI Overview citation and grounding monitoring is justified. The observed source-domain difference does not prove a separate Google index or disclose internal ranking rules.
What remains uncertain
How do these study-specific rates change by region, device, topic, personalization and later Google releases? How do classifier and page-fetch limitations affect the estimated unsupported-claim rate?