Google’s new artificial intelligence is considered the world’s best at weather forecasting

Google’s DeepMind artificial intelligence (AI) laboratory has introduced a new model that is currently considered the best at what it does. This GenCastA probability platform with impressive results in weather forecasting and identifying weather events.

According to DeepMind’s announcement, artificial intelligence can: To ensure the possibility of certain conditions up to 15 days in advance It has a higher accuracy rate than the European Center for Medium-Range Weather Forecasts, a respected organization in the region.

DeepMind’s equipment in total It can make 15-day forecasts in up to 8 minutes.. Google still limited to labs will release all model code to the community The meteorology industry has an expectation of developing new systems based on GenCast.

What does GenCast do differently?

The difference of the new system is in the results it provides. Instead of a single weather forecast typical among other weather tools, GenCast Acts based on the likelihood of 50 or more simultaneous scenariosIt provides a prediction for each event. Calculated items include: Temperature, wind speed and pressure at different altitudes.

Powered by model updated climate data as well as events from previous decades Also to identify patterns and changes according to the time of year. When it receives an order, it compares random parameters with previous results to generate the probability.

In this way, in addition to the conditions of a day, at the same time Can predict extreme weather events such as hurricanes and storms. Such warning is vital so that governments have time to organize themselves, allocate resources and prepare the population for possible problems.

The data can be used in the energy sector as well as in weather forecasts to prevent natural disasters. for example to calculate the operation of wind turbinesIt is based on the winds of a period. Google now also plans to work collaboratively with meteorologists to further develop the model and find practical applications for the technology.