Natural disasters pose a serious threat to human life and sustainable development. The advancement of artificial intelligence (AI) technology opens new opportunities for more effective and efficient disaster management. This article aims to systematically review the use of AI in various phases of disaster management, from mitigation, preparedness, emergency response, to post-disaster recovery. Through a PRISMA-based literature review of 20 Scopus Q1-Q2 and SINTA-indexed articles (2020-2025), this study finds that AI including machine learning, deep learning, Generative AI, NLP, and IoT has significantly contributed to improving prediction accuracy, detection speed, effectiveness of Early Warning Systems (EWS), and disaster response coordination. Key findings include: (1) 15.61% annual growth in AI-disaster research; (2) dominance of application in the mitigation phase; (3) large implementation gaps in developing countries; and (4) urgency of developing transparent AI (Explainable AI/XAI). This article contributes to the development of a theoretical and empirical framework for AI adoption in disaster governance in Indonesia.
Copyrights © 2026