US EditionPublished • loading... • Updated
Google Search and Gemini Users to Get More Accurate, Timely Weather Forecasts Powered by AI
WeatherNext 2 forecasts weather variables 8 times faster with 99.9% accuracy and predicts hundreds of possible outcomes including rare catastrophic events, Google said.
- Today’s date is 2025-11-18, Google DeepMind and Google Research announced WeatherNext 2 as their most advanced and efficient forecasting model, integrated into Google's core system and products like Pixel Weather, Search, and Gemini, with cloud access via Google Cloud services.
- To deliver faster, higher-resolution forecasts, WeatherNext 2 was developed to run eight times faster and reach 1-hour resolution, producing forecasts up to 15 days ahead and serving enterprise customers and consumers.
- Technically, WeatherNext 2 relies on a Functional Generative Network that injects targeted randomness to produce many outcomes from marginals, forecasts complex joints, and runs ensembles in under a minute on a single TPU.
- Early access and cloud data tools let customers run custom models and perform geospatial or large-scale analysis, with industries like energy and agriculture using one-hour step forecasts for precise decisions.
- Amid competition from ECMWF, Nvidia and Huawei, Google plans to integrate new data sources and expand WeatherNext 2 ensemble forecasts to improve preparedness for low-probability catastrophic events.
Insights by Ground AI
Podcasts & Opinions
31 Articles
31 Articles
Google updates its weather forecasts with a new AI model
An illustration from Google’s new AI weather model. | Image: Google Google debuted a new and improved AI weather model today that will be incorporated into popular products like Search, Gemini, and Pixel phones. The company’s AI-enhanced forecasts have proven themselves to be quite accurate so far. AI weather models have also been able to make predictions faster and more efficiently than conventional physics-based models. Until now, Google’s w…
·United States
Read Full Article
Coverage Details
Total News Sources31
Leaning Left2Leaning Right1Center3Last UpdatedBias Distribution50% Center
Bias Distribution
- 50% of the sources are Center
50% Center
L 33%
C 50%
R 17%
Factuality
To view factuality data please Upgrade to Premium
