AI models are being said to be very helpful in predicting the weather. In this direction, Google's DeMind researchers have introduced a new AI model. This model is said to be 30 percent more accurate than the predictions made in the traditional way. This model called Gencast can accurately predict not only the normal weather but also thunderstorms and extreme cold-heat events.

One special thing about it is that the new AI model can complete its predictions in just 8 minutes. Whereas the traditional system requires supercomputers to run, which takes more time.

Leaving the European Center behind in 97.2 percent of cases

When GenCast was compared to the European agency for medium range weather forecasting, it proved to be more accurate in 97.2 percent of the cases, especially when the weather was forecast for more than 36 hours, the accuracy of GenCast proved to be 99.8 percent correct. The model was trained on temperature, wind speed and air pressure data for four decades from 1979 to 2018.

What is Zencast?

Gencast is part of Google's growing suite of next-generation AI-based weather models, including Google DeepMind's AI-based deterministic medium-range forecasting, and Google Research's NeuralGCM, SEEDS, and flood models.