Culture & History

Share This Post

Global Public Data-AI Partners

Share This Post

2026.01.13
CultureAI
0 0
  • Generative AI platforms worldwide are increasingly being used to share information about cultural heritage sites, historical narratives, and national identities. However, these systems frequently produce distorted, incomplete, or factually incorrect descriptions of non-Western cultural heritage—including Korean and Gyeonggi Province landmarks such as Suwon Hwaseong Fortress and Namsadang Nori.

    These errors are not merely technical glitches. They reflect a structural imbalance: AI models are predominantly trained on datasets that lack diverse, verified, and structured public data from national and local governments outside major Western countries. This results in cultural misrepresentation and undermines global trust in AI-generated knowledge.

    We propose the establishment of an international framework to ensure that AI systems accurately represent cultural heritage worldwide:

    1. UNESCO-led 'Global Public Data for AI' Initiative
    - Create an open repository of UNESCO-designated heritage sites with verified descriptions, images, historical context, and structured metadata.
    - Encourage member states to contribute standardized public datasets (JSON, RDF formats) for AI training.
    - Establish guidelines for ethical AI training on cultural content to prevent distortion and appropriation.

    2. AI Platform Accountability Framework
    - Major generative AI platforms (OpenAI, Google, Microsoft, etc.) should commit to integrating verified public datasets from UNESCO and national cultural institutions.
    - Platforms must establish transparent processes for correcting cultural errors reported by citizens and experts.
    - Implement periodic audits to ensure cultural heritage information aligns with official sources.

    3. Citizen-Led Global Verification Network
    - Support international digital ambassadors (like VANK's Global AI Ambassadors) who identify AI errors and propose corrections based on public data.
    - Develop multilingual reporting platforms where users can flag cultural misrepresentation in AI outputs.
    - Share best practices across countries to build a collective knowledge base for AI accuracy.

    4. Cross-Border AI Ethics Education
    - Integrate "AI and Cultural Integrity" modules into computer science and data science curricula globally.
    - Promote collaboration between software engineers, historians, and cultural experts to design culturally responsible AI systems.

    Cultural heritage belongs to humanity, and its representation in AI systems must be accurate, respectful, and inclusive. This proposal is not just about correcting data—it is about ensuring that AI serves as a bridge for global understanding, not a source of cultural distortion.

    We call on UNESCO, national governments, AI companies, and global citizens to work together in building a trustworthy, culturally responsible AI ecosystem for future generations.

Reference URL 0

Related Petitions

Tags
0 have signed

Add your voice. Sign the petition.

0 have signed

Add your voice. Sign the petition.