Aligning Language Models with Indigenous Community Cosmovision through Constitutional AI

Research International Source checked

Source: arXiv (Cornell University)
Published: 18 Oct 2026
Authors: Silvia Fernandez-Sabido, Jorge Reyes-Magaña, Jesús Adrián Chay Tun, Jean Carlos Buenfil, Victor González Nahuat, Francisco Morales

Excerpt from source

Large language models (LLMs) are predominantly trained on corpora that reflect Western, anglophone worldviews. This asymmetry is not neutral: when a model answers questions about territory, community authority, or the relationship with nature, it does so from assumptions that may be alien to the communities using the system. This paper presents a system for aligning language models with the ethical principles of Indigenous communities in Latin America. Drawing on interviews with five Indigenous leaders and defenders, we built an operational cultural constitution grounded in an Indigenous cosmovision, with 12 principles. On this basis, we implemented a Constitutional AI (CAI) pipeline in two…

Read the source ↗

These stories are gathered automatically from across the web and each source link is checked. A summary never replaces the original — always read the source. Inclusion here is not endorsement.

Related stories

Phenotypic Evaluation of Sunflower (Helianthus Annuus L.) Landraces as a Tool for Conservation and Valorization of Plant Genetic Resources

Community-Governed Digital Preservation of Zambezi Indigenous Knowledge: Transitioning Oral Traditions to Sustainable Learning Documentation in Namibia

Development and implementation of an Indigenous data governance framework and stewardship tool to document, store, and apply Indigenous environmental data in northern Canada

In Order to ‘Say What We Say:’ Archival Protocol that Attends to Indigenous Data Sovereignty