Google's Weather Forecasting Model Just Got a Massive Upgrade
- On Thursday, Google DeepMind introduced WeatherNext 3, an AI weather forecasting model generating hourly predictions using live satellite imagery instead of government datasets that update every six hours.
- Traditional weather models relied on complex physics simulations updating every six hours, creating data lags for fast-changing variables like rain; the new system uses real-time observations to provide a global picture roughly five times sharper than WeatherNext 2.
- Designed for precision, the system forecasts wind speeds at 100 meters, aiding renewable energy developers in estimating power output, while achieving up to 60% better precipitation accuracy compared to previous versions.
- Google is immediately integrating these forecasts into Search, Maps, and the Gemini app, offering users planning a day or more ahead up to 50% more accurate precipitation predictions.
- The platform brings high-fidelity forecasting to Latin America, Africa, and Asia-Pacific, regions historically underserved by accurate forecasts; Bill Gates recently cited AI-powered weather forecasting as crucial for improving crop yields in developing countries.
26 Articles
26 Articles
Google says its new AI weather system can talk to satellites and deliver clear forecasts every hour
Google has launched WeatherNext 3, an AI weather model that can deliver forecasts every hour at up to 5km resolution using real-time satellite data. The model also promises better rain forecasts and is coming to Google Search, Gemini, Maps and other Google products.
The new system uses real-time satellite data instead of traditional physical weather forecasting models.
Google's WeatherNext 3 AI Model Harnesses Live Satellite Data to Sharpen Global Forecasts
Google DeepMind and Google Research unveiled WeatherNext 3 on Wednesday, a new AI system that ingests real-time satellite imagery to deliver hourly weather predictions at resolutions five times sharper than its predecessor. The model marks a clear break from traditional numerical weather prediction methods that rely on physics simulations running on supercomputers with built-in delays. Most AI weather models train on output from those numerical …
Coverage Details
Bias Distribution
- 72% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
