Recent research from the University of Geneva reveals that artificial intelligence (AI) tools for weather prediction struggle with forecasting extreme temperature and wind events. Analyzing global weather data from 1979 to 2020, scientists found over 162,000 heat records and significant shortcomings in AI’s predictions when compared to traditional physics-based models. While AI has advanced weather forecasting techniques, it often underestimates heat wave intensities and fails to recognize unprecedented extreme conditions. This gap is concerning, as accurate forecasts are critical for managing public safety and resource planning during severe weather. Furthermore, AI systems’ reliance on historical data limits their effectiveness, posing risks to communities. Researchers are seeking solutions, including using synthetic weather scenarios and integrating machine learning with physics models to improve accuracy. Staying informed through local emergency alerts and supporting weather prediction research initiatives can help communities better prepare for unpredictable weather extremes.
Source link
