MIT AI predicts extreme weather without past data

MIT engineers have created an AI tool capable of forecasting extreme weather events without using historical disaster data. Developed by graduate student Kai Chang and Professor Themis Sapsis, the tool, named Extreme Event Aware, uses point statistics and spatial maps to predict events that have not yet occurred but are statistically possible. This method allows for the generation of scenarios such as unprecedented storms, aiding planners and insurers in preparing for rare but severe weather conditions. The research was published in Nature Communications.