Artificial Intelligence (AI): the potential for MHEWS

Over the past year, concerns about the viability of AI gave way to a surge in practical application among organisations (McKinsey, 2024). Yet concerns remain over how to ensure AI works for everyone. UNU outlines a series of potential applications (UN University, 2024): 

• Advance disaster risk knowledge by gathering and analysing vulnerability and exposure data, particularly useful in areas where data is scarce. However, ensuring that AI models don't misinterpret or overlook nuances of marginalised groups is crucial for ensuring people-centred principles.
• Accelerate hazard detection and monitoring through improved predictive analytics and real-time data assessments. WMO's Severe Weather Information Center uses AI to consolidate and disseminate severe weather information. This is reliant on significant resources and data, which may not be available in many countries.
• Improve communication and dissemination by optimising when, where, and how alerts are sent. AI can also translate alerts into multiple languages and customise information to highlight context-specific impact-based alerts. Equally, AI can generate false and misleading information that spreads at speed and scale. This requires stronger policies, professional skills, and media literacy levels to maintain healthy communication ecosystems.
• Simulate emergency scenarios that boost preparedness and response. AI can present complex scenarios based on data, overriding our own limitations that prevent us from imagining certain events and how they may unfold. AI can also aid in preparing for scenarios, refining contingency plans, and allocating resources through real-time assessments. Relying on untested AI methods for decision-making during fast-moving events with real-life consequences requires great care.