FAIA - FAIR AI Attribution White Paper

Zenodo (CERN European Organization for Nuclear Research)

Abstract

This white paper describes the FAIA - FAIR AI Attribution Framework. FAIA provides a vocabulary and technical framework for machine-readable, persistent, and verifiable disclosure of AI involvement. It defines three complementary elements: high-level attribution flags (human-created, AI-assisted, AI-generated), activity codes describing the role AI played in the content lifecycle, and optional system attribution identifying the AI system and version involved. FAIA declarations can be bound to ISCC fingerprints, allowing attribution information to remain resolvable even when files are copied, transformed, or stripped of metadata. FAIA supports consistent transparency across sectors including publishing, journalism, research, and media production. It supports compliance with emerging obligations such as the EU AI Act and provides a foundation for services that depend on reliable provenance information, including content verification, moderation, search, and training data curation.

Authors 5

1. Sylvia E. Le Dévédec Aachen

   Leiden University

   Affiliation as printed

   Leiden University

2. Kristina Maria Hettne Aachen

   Leiden University

   Affiliation as printed

   Leiden University

3. Alessa A. Gambardella Aachen

   Leiden University

   Affiliation as printed

   Leiden University

4. Erik Schultes

   GO FAIR Foundation

   Affiliation as printed

   Leiden Academic Center for Drug Research

   partners in FAIR

   GO FAIR Foundation

5. Sebastian Posth
   Affiliation as printed

   Liccium