Abstract

This review critically examines the transformative role of artificial intelligence (AI) in coordinating the core phases of disaster response: preparedness, response, recovery, and mitigation.  Drawing on illustrative case studies such as AI-driven flood forecasting with Delft-FEWS, post-disaster damage mapping via DroneDeploy, and optimised emergency dispatch through RescueME. It synthesises evidence on five core AI capabilities (machine learning, deep learning, computer vision, natural language processing, and optimisation algorithms). We find that AI can substantially improve predictive accuracy, real-time situational awareness, rapid decision-making, resource allocation, and inter-agency collaboration, thereby addressing the speed and complexity challenges of traditional disaster management. However, adoption is hindered by fragmented data ecosystems, opaque “black box” models, interoperability gaps, cybersecurity vulnerabilities, ethical and equity concerns, and limited accessibility in low-resource settings. To overcome these barriers, we argue for the development of interoperable data standards, explainable AI frameworks, robust cyber governance protocols, and inclusive stakeholder engagement. Emerging trends, such as the convergence of AI with IoT and edge computing, enhanced human-AI decision support, and the democratisation of AI tools, offer promising pathways for building more resilient, scalable, and ethically grounded disaster response systems. By aligning technological innovation with human oversight and participatory governance, strategic integration of AI can enhance preparedness, response effectiveness, and recovery efficiency, fostering safer and more resilient communities worldwide.