ClaimReview Markup

ClaimReview is a specialized Schema.org structured data type designed to annotate a specific claim with a corresponding review and rating. It extends the base Review type to include a claimReviewed property, which contains the exact statement being fact-checked. The markup must be embedded as JSON-LD within a <script> tag on the page where the fact-check appears. Search engines parse this structured data to generate rich results, including the claim text, the publisher's rating, and a link to the full analysis.

Google requires specific properties for a ClaimReview to be eligible for display as a fact-check rich result:

• claimReviewed: The exact, verbatim text of the claim being evaluated.
• author: The organization or person publishing the fact-check, with a name and url.
• reviewRating: A nested Rating object containing ratingValue (e.g., 'True', 'False', 'Mostly False') and bestRating/worstRating.
• url: The canonical URL of the full fact-check article.
• itemReviewed: A CreativeWork describing the source of the claim (e.g., a social media post or speech).

The reviewRating property uses a standardized numeric scale to normalize diverse publisher rating systems:

• ratingValue: A numeric value representing the truthfulness (e.g., 1 for false, 5 for true).
• bestRating: The maximum value on the scale (typically 5).
• worstRating: The minimum value on the scale (typically 1).
• alternateName: A human-readable label like 'Pants on Fire' or 'Four Pinocchios'. This normalization allows search engines to aggregate and compare fact-checks across different organizations with varying rating taxonomies.

When properly implemented, ClaimReview markup enables a fact-check rich result in Google Search. This snippet displays:

• The claim text in quotation marks.
• The rating (e.g., 'False' with a visual indicator).
• The publisher name and publication date.
• A direct link to the full fact-check article. These rich results appear in Google News, Search, and Google Images. They are also surfaced in Google Assistant responses and can be ingested by third-party AI systems for claim verification tasks.

Not all publishers can have their ClaimReview markup displayed. Google enforces strict eligibility criteria:

• The publisher must be a verified signatory of the International Fact-Checking Network (IFCN) Code of Principles or an equivalent recognized body.
• The fact-check must analyze a discrete, verifiable claim about a specific factual matter.
• Opinion pieces, political commentary, and general news analysis do not qualify.
• The markup must appear on a page that contains the full fact-check methodology and sources, not just a summary.

ClaimReview markup contributes to entity identity and trust signals within Google's Knowledge Graph. By consistently publishing verified fact-checks with structured data, an organization reinforces its Organization entity as an authoritative source. The author property links the fact-check to a specific entity URI, building a graph of verified assertions. This semantic linkage supports broader Generative Engine Optimization strategies by establishing the publisher as a high-confidence node for AI-driven question answering and claim verification pipelines.

The automated process of assessing the truthfulness of a textual claim by corroborating it against a trusted knowledge base or evidence corpus. Fact verification systems use natural language inference and evidence retrieval to classify claims as true, false, or mixed. ClaimReview markup is the structured data output layer that communicates these verification results to search engines and AI crawlers.

An NLP task that identifies textual mentions of entities and maps them to unique, unambiguous entries in a target knowledge base like Wikidata or DBpedia. For ClaimReview to be effective, the subject of a claim must be precisely disambiguated:

• Links the claim subject to a canonical URI
• Prevents confusion between similarly named entities
• Enables search engines to aggregate fact-checks about the same entity across sources

The strategic implementation of Schema.org vocabulary to define entities, attributes, and relationships for AI-driven search engines. ClaimReview is a specific Schema.org type within the broader CreativeWork hierarchy. Proper implementation requires:

• Valid JSON-LD serialization embedded in HTML
• Correct nesting of itemReviewed and author properties
• Accurate reviewRating values using the defined rating scale

Metadata that tracks the origin, source, and transformation history of a specific fact or entity within a knowledge graph. For ClaimReview, provenance is critical:

• Documents the original source of the claim being reviewed
• Establishes the authority of the fact-checking organization
• Provides an auditable trail for data lineage and trust assessment by AI models

Techniques for embedding explicit markers of certainty, source quality, and data freshness within content to guide an AI model's trust assessment. ClaimReview markup serves as a high-confidence signal by:

• Providing a structured, machine-readable verdict
• Linking to the full fact-check article for verification
• Including the datePublished to indicate freshness of the review

The specific sub-task of entity linking that resolves which distinct real-world entity a textual mention refers to when the name is ambiguous. For example, distinguishing 'Paris' the city from 'Paris' the mythological figure. In ClaimReview, precise disambiguation ensures that fact-checks are correctly associated with the intended subject entity, preventing misattribution in search results and knowledge panels.