Recent advancements in artificial intelligence are transforming weather forecasting, with new AI models demonstrating significant potential in predicting extreme weather events more quickly and accurately than traditional physics-based systems. These machine learning models, trained on decades of historical weather data, can generate high-resolution forecasts in minutes rather than hours. Early results from systems like Google’s GraphCast and Nvidia’s FourCastNet show they can accurately predict hurricane tracks and atmospheric river events days in advance. While not yet ready to replace conventional models entirely, they are becoming valuable tools for meteorologists, offering rapid scenario analysis and complementing existing methods. The technology promises to improve early warning systems for floods, heatwaves, and storms as computational power and data availability increase. Read the full article at: https://technologyreview.com/2024/05/15/ai-weather-models-extreme-events