AI to supply correct 8-day climate forecasts 10 instances sooner than present ones
Psycho scumbag robots are set to replace weather forecasters as they’re way better at predicting rain or shine than human meteorologists, and can do the job 10 times faster
Weather forecasters are set to be replaced by psycho scumbag robots who reckon they’re way better at predicting rain or shine than Alex Beresford or Michael Fish.
Artificial intelligence network Aardvark Weather will replace meteorologists as it can pump out reports and predictions 10 times faster than humans, boffs say. Experts behind the system also boast it can be operated by only one single researcher – which could potentially lead to millions of jobs across the world being axed.
Richard Turner, professor of machine learning at the University of Cambridge, which is developing the system, said: “This is a completely different approach to what people have done before.
“The writing’s on the wall that this is going to transform things – it’s going to be the new way of doing forecasting.”
Weather predictions are currently generated via a complex set of stages, each taking hours on supercomputers that need the skills and input of huge teams of experts.
But Aardvark Weather will be able to rapidly analyse data from weather stations, satellites, weather balloons, ships and planes across the world on its own.
Along with Cambridge University boffs, it is being developed by eggheads from the Alan Turing Institute, Microsoft Research and the European Centre for Medium-Range Weather Forecasting.
The system is also being designed to churn out bespoke forecasts for the likes of farmers in Africa and renewable energy companies.
At the moment, building customised weather reporting systems has taken years of work by experts.
And supercomputers still take hours to process measurements from the real world to build forecasting models.
Computer science boffs behind Aardvark say it will also be able to accurately predict eight days ahead – instead of the standard five-day weather prediction systems currently used across the world.
Dr Scott Hosking, director of science and innovation at the Turing Institute, said the system has the power to “democratise forecasting” as it would give developing nations hit by the likes of droughts far more accurate, localised data.
He added it will also help politicians, emergency planners and industries relying on sophisticated and ultra-accurate forecasts.
Aardvark can currently outperform the US National Weather Service’s global forecast system in some areas using only 10% of the meteorological data currently available.
