A new artificial intelligence weather forecasting model developed by Google DeepMind, named GraphCast, has demonstrated superior performance compared to the world’s leading conventional system. The model, described in a Science journal paper, can generate accurate 10-day global weather predictions in under one minute on a single machine. In rigorous testing against the European Centre for Medium-Range Weather Forecasts (ECMWF) system, GraphCast outperformed it on over 90% of 1,380 metrics, including critical predictions for extreme weather events like tropical cyclones and atmospheric rivers. The AI system was trained on nearly four decades of historical weather data, learning to model complex atmospheric physics. While not intended to replace traditional methods, it represents a significant leap in speed and cost-efficiency for weather prediction, offering a powerful complementary tool for meteorologists. Read the full article at https://technologyreview.com/2023/11/14/1084081/ai-weather-prediction-model-beats-traditional-forecasts/.