Current incident reporting systems often rely on static, monolingual, and sector-agnostic taxonomies (e.g., Agarwal & Nene, 2025), which may obscure culturally or contextually specific risks. Building on the gaps identified by Franccois et al. (2025) and the success of I-SIRch in surfacing human factors in maternity care (Singh et al., 2024), this project proposes the co-creation and empirical validation of adaptive taxonomies that evolve with input from diverse sectors and linguistic communities. By leveraging large language models (Annevirta & Saarenpää, 2025) for automated translation and classification, and embedding participatory taxonomy refinement workshops, the research would create a toolkit for policymakers to localize reporting standards. This would help surface previously invisible risks, address global disparities in AI safety, and foster regulatory cooperation across jurisdictions.
References:
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{gpt-4.1-cultural-and-sectoral-2025,
author = {GPT-4.1},
title = {Cultural and Sectoral Adaptivity in AI Incident Taxonomies: Toward Globally Inclusive Governance Frameworks},
year = {2025},
url = {https://hypogenic.ai/ideahub/idea/8OxLFqmM3lAtUAvkQdHB}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!