A new study demonstrates that artificial intelligence models can generate accurate weather forecasts faster and at lower computational cost than conventional physics-based systems. The research, led by scientists at the University of California, Berkeley, tested an AI system that was trained on decades of historical weather data. It successfully predicted global atmospheric conditions up to ten days in advance with accuracy comparable to, and in some cases exceeding, the current gold-standard European model. The AI system completed its forecasts in under a minute on a single desktop computer, a task that typically requires hours on a supercomputer for traditional methods. While experts note that AI models may still struggle with predicting extreme, rare weather events due to limited training data, the technology represents a significant shift toward more efficient and accessible forecasting. For the full details, read the complete article at the provided source link.