A new artificial intelligence 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 system, named GraphCast, uses machine learning trained on decades of historical weather data to predict atmospheric conditions up to 10 days in advance. In head-to-head tests against the European Centre for Medium-Range Weather Forecasts’ high-resolution system (HRES), GraphCast provided more accurate predictions for over 90% of 1,380 test variables, including critical measurements like temperature, pressure, and wind speed. The AI model generates a 10-day forecast in under a minute on a single Google TPU v4 machine, a process that takes traditional supercomputers hours to complete. Researchers emphasize that AI models like GraphCast are not replacements but powerful complements to existing numerical weather prediction systems, offering faster, more efficient forecasts that could improve severe weather warnings. The model’s code has been made open-source to encourage further development and integration by meteorological agencies worldwide. Read the full article at: https://technologyreview.com/2024/11/14/ai-weather-model-graphcast-outperforms-traditional-forecasts/