Google has launched a new AI-powered weather app called GenCast, which claims to deliver faster and more accurate forecasts up to 15 days ahead. The model outperforms one of the world’s best forecast models, ENS, with an impressive accuracy rate of 97.2% in predicting weather conditions for the year 2019.
GenCast was trained on four decades worth of weather data and uses AI diffusion technology to recognize patterns and make predictions about future weather events. The model excels at speed, producing a 15-day forecast in just eight minutes compared to traditional models that can take several hours.
According to Matt Chantry, ECMWF machine learning coordinator, GenCast marks a significant milestone in the evolution of weather forecasting. Ilan Price, senior research scientist at DeepMind, notes that traditional forecasts are computationally expensive and less efficient than GenCast.
While it remains to be seen whether AI-powered weather models will become useful in real-world applications, better longer-term forecasts would benefit everyone, not just photographers. The code for GenCast is now open and available for download, offering a promising solution for improving weather forecasting accuracy.
Source: https://petapixel.com/2024/12/09/googles-ai-weather-app-could-be-extremely-useful-to-photographers