DeepMind Unveils WeatherNext 2, Boosting AI Weather Forecast Accuracy
Key Points
- DeepMind launches WeatherNext 2, an AI model for global weather forecasting.
- Provides up to two‑week forecasts with hourly resolution.
- Improves tropical‑storm track predictions to three days ahead.
- Runs eight times faster than the previous version.
- Offers higher‑resolution data for industries like energy trading.
- Acknowledges challenges with outlier rain and snow events.
- AI weather forecasting is a growing competitive arena.
Google's DeepMind has launched WeatherNext 2, an advanced AI weather‑prediction model that delivers faster, higher‑resolution forecasts up to two weeks ahead. The new system improves tropical‑storm track predictions, offers hourly updates, and runs eight times faster than its predecessor, promising more precise decisions for industries such as energy trading. While the model still struggles with outlier rain and snow events, DeepMind says ongoing research will address these gaps as AI continues to reshape meteorology.
Model Overview
DeepMind announced the release of WeatherNext 2, the latest iteration of its artificial‑intelligence weather forecasting platform. Built on a novel approach that replaces older image‑and‑video‑generation models with a single‑step processing architecture, WeatherNext 2 reduces reliance on costly AI compute resources while delivering more granular forecasts.
Performance Gains
The company claims the new model "delivers more efficient, more accurate and higher‑resolution global weather predictions." It can now provide forecasts up to two weeks out, covering temperature, pressure, and wind. In tropical‑storm scenarios, WeatherNext 2 improves track predictions, offering reliable guidance up to three days ahead—an advance over the prior model’s two‑day horizon. Hourly forecasts are also incorporated, giving users finer‑grained insight. DeepMind reports the system runs eight times faster than its predecessor, a speed boost that helps businesses act on rapidly changing conditions.
Industry Impact
According to DeepMind researchers, the higher resolution and speed enable sectors like energy trading to make more precise decisions, enhancing resilience to weather variability. "It gives you a more granular forecast," said AI researcher Akib Uddin. "Many other industries are quite interested in these one‑hour steps. It helps them make more precise decisions. Their goal is, how can they make their business more resilient to weather?"
Limitations and Ongoing Work
Despite its advances, DeepMind acknowledges that WeatherNext 2 may still miss outlier rain and snow events due to gaps in training data. "It’s one limitation of our forecast, but one that we are working on improving," explained research scientist Ferran Alet. The company is actively refining the model to address these shortcomings.
Competitive Landscape
The AI weather‑prediction field is becoming increasingly crowded. Companies such as NVIDIA, Microsoft, AccuWeather, and Huawei are also developing their own solutions, intensifying competition and driving rapid innovation across the sector.