A new artificial intelligence weather forecasting model developed by Google DeepMind has demonstrated superior accuracy compared to traditional physics-based systems in global medium-range weather predictions. The model, named GraphCast, uses machine learning to analyze decades of historical weather data and identify complex atmospheric patterns. In head-to-head tests against the European Centre for Medium-Range Weather Forecasts’ high-resolution forecast system, GraphCast provided more accurate predictions for over 90% of 1,380 test variables, including critical measurements like atmospheric pressure, temperature, and wind speed. The AI system can generate a 10-day forecast in under a minute on a single Google TPU, significantly faster than conventional methods that require hours of supercomputer calculations. Researchers emphasize that AI models like GraphCast should complement rather than replace traditional systems, as they rely on established models for training data and lack the deep physical understanding of atmospheric science. The technology represents a significant shift toward data-driven weather prediction that could improve early warnings for extreme weather events. Read the full article at https://technologyreview.com/2024/01/15/ai-weather-model-outperforms-traditional-forecasts.