disaster risk management

This study examines the intersection of artificial intelligence, global governance, and disaster risk management through a qualitative investigation of 92 Iranian experts across disciplines such as geoinformatics, emergency planning, and environmental engineering. While AI offers a significant promise for enhancing early warning systems, damage assessments, and real-time decision-making, its integration into DRM systems remains constrained by fragmented data infrastructures, institutional silos, and geopolitical exclusions. Participants underscored AI’s potential to improve response coordination and risk forecasting, but emphasized the need for robust data governance, algorithmic transparency, and capacity building. The study highlights critical ethical and political concerns—particularly in countries like Iran facing technological marginalization due to sanctions and limited access to global data ecosystems. Drawing on grounded theory and thematic analysis, the research identifies institutional fragmentation, interoperability barriers, and normative governance deficits as primary obstacles to AI-enabled DRM. It argues for a globally coordinated approach grounded in justice, inclusivity, and human-centered design.

Background: This study explores the paradoxical role of artificial intelligence as both a tool of exploitation within surveillance capitalism and a force for resilience in disaster risk management, highlighting the ethical and governance challenges that arise at the intersection of these domains. Aims: This article aims to enable the ethical use of AI in DRM while insulating public systems from the structural harms of commercial data exploitation. Methodology: Drawing from a comparative qualitative analysis of 35 academic sources, the study investigates how BigTech corporations and data brokers leverage AI to commodify personal data, consolidate informational power, and erode democratic agency. Results: The present study critically examines the dual role of artificial intelligence in contemporary digital society, contrasting its exploitative deployment within surveillance capitalism with its constructive use in disaster risk management (DRM). Simultaneously, it highlights a parallel body of research showcasing AI’s capacity to enhance early warning systems, situational awareness, and post-disaster recovery, especially in resource-constrained and climate-vulnerable regions. Conclusion: To reconcile these conflicting trajectories, the article proposes the Public AI for Resilience (PAIR) framework—a governance model grounded in data sovereignty, public infrastructure, algorithmic explainability, and cross-sectoral collaboration. Ultimately, the article argues for a normative recalibration of AI governance that prioritizes equity, transparency, and collective resilience over market imperatives, demonstrating that AI’s public good potential can be realized without surrendering to surveillance-based capitalism.