Schema markup gets an actual number behind the folklore
The debate finally has data
'Does schema.org markup actually help with AI search?' has been argued in SEO forums for years, mostly without hard numbers attached to either side.
That's starting to change: pages carrying proper schema markup now show roughly a 2.5x higher chance of appearing in AI-generated answers than comparable pages without it, and sites with a complete author/date/FAQ-tier of markup see up to 40% more AI Overview appearances.
What 'not required' really means
Google's own guidance says no special markup is required for AI Overviews or AI Mode specifically — true in the narrow sense that the feature doesn't have a documented schema requirement the way, say, recipe rich results do.
What schema actually does is different and arguably more important: it helps a language model correctly and quickly identify what a page is about, who wrote it, and how it relates to known entities, before the model ever decides whether to trust and quote it. 'Not required' and 'doesn't help' are two very different claims, and the data lines up with the second one being false.
The schema worth prioritizing
In practice, the schema worth prioritizing for AI visibility looks a lot like the schema that was already good practice: author and organization markup so a model can attribute the content to someone real, datePublished/dateModified so freshness is unambiguous, and FAQPage/HowTo or AggregateRating where the content genuinely fits that shape.
It's a cheap, one-time technical fix — which is exactly the kind of signal an AI Readiness check flags directly, since it's checking for the same schema.org fields.
Based on reporting from Layer3 Labs
Is your own content set up for stories like these?
Run a free AI Readiness check — retrieval, extractability, schema.org signals, and a prioritized rewrite brief, scored the way an AI assistant actually reads your page.