Google says its new AI model outperforms the top weather forecast system | TechCrunch
Dec 07, 2024 - techcrunch.com
Google's DeepMind team has introduced an AI model named GenCast for weather prediction, which reportedly surpasses the European Centre for Medium-Range Weather Forecasts’ ENS, considered the leading operational forecasting system. According to a paper published in Nature, GenCast was trained on weather data up to 2018 and demonstrated higher accuracy in forecasts for 2019, outperforming ENS 97.2 percent of the time. Unlike its predecessor, which offered a single deterministic forecast, GenCast generates an ensemble of over 50 predictions, providing a complex probability distribution of potential weather scenarios.
Google plans to integrate GenCast into its suite of AI-based weather models, enhancing services like Google Search and Maps. Additionally, the company intends to make GenCast's real-time and historical forecasts publicly available for use in research and other models. This development marks a significant advancement in AI-driven weather forecasting, offering more accurate and probabilistic predictions.
Key takeaways:
```html
DeepMind's new AI model, GenCast, has been unveiled for weather prediction.
GenCast outperforms the European Centre for Medium-Range Weather Forecasts’ ENS, the leading operational forecasting system.
GenCast uses an ensemble of 50 or more predictions to create a complex probability distribution of future weather scenarios.
Google plans to incorporate GenCast into Google Search and Maps, and release real-time and historical forecasts for public use.
