A new artificial intelligence-based weather forecasting model developed by researchers at Google DeepMind has demonstrated superior accuracy compared to conventional physics-based models in global medium-range weather predictions. The AI system, named GraphCast, utilizes 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 (ECMWF) system—considered the global gold standard—GraphCast provided more accurate predictions for over 90% of 1,380 test variables, including critical metrics like temperature, pressure, wind speed, and humidity at multiple atmospheric levels. The model operates significantly faster, generating a 10-day forecast in under one minute on a single Google TPU v4 machine, compared to hours of computation on supercomputers required by traditional methods. While the AI model excels at pattern recognition from vast datasets, experts note that traditional systems remain crucial for their grounding in physical laws and ability to simulate unprecedented events. The development marks a significant shift toward data-driven, AI-enhanced weather prediction. Read the full article at: https://technologyreview.com/2024/03/14/ai-weather-model-graphcast-outperforms-traditional-forecasts