Fact Check Checker (FCC): A Structure-First Protocol for Auditing Fact-Check Validity
Abstract
A structure-first protocol to audit whether a fact check has correctly identified what should be verified. This project reconstructs what was meant from primary sources and compares it with what was reported as fact. It detects structural mislabeling such as metaphor-to-fact conversion, condition stripping, scope loss, and subject substitution. Fact Check Checker does not judge speakers, infer intent, or take political positions. It produces a transparent audit result and a mislabeling heat score that quantifies structural drift—before any judgment is made. Included: Formal specification (spec.json) Model-agnostic runner prompt Worked examples (political and non-political) Minimal ASCII flow License: CC0 / CC BY 4.0 From “what was reported as said” to “what was meant.”
Fields of Study & Subjects
Misinformation and Its Impacts
Ethics and Social Impacts of AI
Humor Studies and Applications
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 15 January 2026