New research published in the open-access Journal of Climate and Atmospheric Science by researchers from the University of Reading reveals that AI-based machine learning can offer faster, less costly and more accurate predictions for major storm paths and damage intensities. The study shows progress and future potential in using such technologies to support monitoring and impact forecasting, to support preparedness and response. However, noted limitations include underestimating wind speed and predictive limitations relating to temperature contrasts within the storm. The researchers call for more rapid research to improve these models.