{
  "id": 5349971,
  "title": "Google’s latest AI weather model gives you no excuse to forget your umbrella",
  "url": "https://urgent.news/2026/09/03/googles-latest-ai-weather-model-gives-you-no-excuse-to-forget-your",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-03T15:00:00.000Z",
  "source": {
    "name": "TechCrunch",
    "slug": "techcrunch",
    "url": "https://techcrunch.com/2026/09/03/googles-latest-ai-weather-model-gives-you-no-excuse-to-forget-your-umbrella/"
  },
  "original_language": "en",
  "account": "Google Deepmind and Google Research have unveiled an advanced artificial intelligence model for weather forecasting called WeatherNext 3. This latest innovation in meteorology, driven by deep learning techniques, promises to revolutionize the way we understand and predict our changing atmosphere. WeatherNext 3 will soon be integrated into various Google products, including search results, Google Maps, and Gemini, as well as accessible through the company's cloud platforms.\n\nThe new model has already demonstrated superior accuracy in comparison to other leading deep-learning models, including those developed by Google, Microsoft, Nvidia, and the European Center for Medium-Range Weather Forecasting. It also outperforms traditional weather forecasts from the US National Weather Service and the ECMWF. Most weather forecasts are generated by government-owned supercomputers, which, while increasingly accurate, are both expensive and slow.\n\nDeepMind staff research scientist manager Ferran Alet explains that machine learning targets the core issue of weather forecasting by learning patterns from vast amounts of data, effectively handling the chaotic nature of the atmosphere and making predictions more quickly with comparable accuracy to government tools. WeatherNext 3 takes on three key challenges faced by AI forecasting models: its ability to predict weather down to a resolution of 5km, improved rain forecasting, and the capability to produce hourly forecasts instead of the standard six-hour predictions.\n\nThe model's enhancements stem from deliberate design choices, such as increasing the number of parameters (2.4 times more than its predecessor) and tailoring the targets for decoder heads to generate more useful answers. While traditional weather forecasts output metrics averaged across a 3D grid, DeepMind researchers have already received praise for tuning their model to visualize cyclone paths. Moreover, WeatherNext 3 now tailors forecasts to specific weather data stations, enabling more granular predictions and facilitating comparisons against ground-truth data.\n\nThis AI model's ability to forecast more frequently stems from its capacity to ingest real-time weather satellite data on an hourly basis. By incorporating raw observations rather than analysis produced by weather supercomputers, WeatherNext 3 promises a more accurate forecast. However, it still relies on national weather datasets, necessitating further work for true direct data assimilation.\n\nGoogle acknowledges that while large language models (LLMs) have garnered significant attention, the transformer revolution in meteorology has also played a crucial role. European and US weather agencies have already incorporated AI models into their forecast products, and their speed and low cost promise to bring economic benefits to regions where high-quality sensors and supercomputers have previously been inaccessible. DeepMind researcher Bill Gates highlighted AI-powered weather forecasting as a vital advantage of the technology, improving crop yields in developing countries.",
  "summary": "Scientists at Google Deepmind and Google Research released a new artificial intelligence model for weather forecasting today that sees our changing atmosphere more clearly and predicts its behavior more often. WeatherNext 3 is the latest wave of a sea change in meteorology brought out by deep learning techniques, and Google says it will start feeding […]",
  "key_points": [
    "WeatherNext 3 AI model enhances weather forecasting accuracy",
    "Model integrated into Google products and cloud platforms",
    "Improved rain forecasting and hourly predictions"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "The Verge",
        "title": "Google says its AI weather model is getting better",
        "url": "https://urgent.news/2026/09/03/google-says-its-ai-weather-model-is-getting-better",
        "published": "2026-09-03T15:00:00.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}