High-stakes decision-making and the future of AI governance: Indigenous self-determination and lessons from the Arctic Council
Loading...
Date
29/04/2026
Authors
Kavanagh, Elena
Journal Title
Journal ISSN
Volume Title
Publisher
Informa UK Limited
Published Version
Abstract
Current AI governance risks remaining dominated by state-centric models that marginalise Indigenous peoples and historically excluded communities. This paper proposes a human-centric AI governance framework that reconceptualises AI governance institutions as socio-technical decision systems responsible for structuring ethical, political, and regulatory choices about AI deployment. It argues that Indigenous self-determination and participatory rights should be embedded as design principles of AI governance architecture rather than treated as consultative add-ons. By integrating relational knowledge systems and collective responsibility models, pluralistic governance structures can enhance the legitimacy, resilience, and ethical robustness of AI-related decision processes. The analysis uses the Arctic Council’s model as an institutional reference to show how co-governance arrangements can inform the design of inclusive AI governance. The paper contributes to decision systems research by extending the concept of the socio-technical decision system to high-stakes AI governance and by formulating institutional principles for pluralistic AI governance, organised around decision-systems logic.
Description
Keywords
Human-centric AI governance , AI governance , Socio-technical decision systems , Indigenous knowledge , Participatory governance , Legal pluralism , Decolonial governance
Citation
Kavanagh, E. (2026) 'High-stakes decision-making and the future of AI governance: Indigenous self-determination and lessons from the Arctic Council', Journal of Decision Systems, 35(1), 2662348 (9pp). https://doi.org/10.1080/12460125.2026.2662348
Link to publisher’s version
Collections
Copyright
© 2026, The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is notaltered, transformed, or built upon in any way. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by theauthor(s) or with their consent.