Google Launches WeatherNext v3, Adds Satellite Data to Cut Forecast Lag

Google has released version 3 of its WeatherNext AI weather model, the first major update in several years. The new model now ingests satellite weather data, a change that shortens the lag between the latest observed conditions and the generation of a fresh forecast. By incorporating real‑time satellite imagery, WeatherNext can produce updated predictions more quickly than before, while still matching the accuracy of traditional numerical models. The update is detailed in a Google white paper and follows the company’s strategy of using AI to deliver comparable forecast performance with less computing power, enabling more frequent runs. The move also highlights the growing role of reanalysis techniques—combining diverse data streams into a consistent global atmospheric snapshot—to support AI‑based forecasting.