Will AI Change The Future of Forecasting Improving Reliability Over Longer Lead-Times?

[Layman]

Thank you @RDRY for posting this interesting Barron's link regarding the GenCast AI weather model:  https://www.barrons.com/news/google-says-ai-weather-model-masters-15-day-forecast-cdc5793d 

Will, or can, AI improve forecasting reliability?  What are your thoughts, opinions and/or experience?  Will there still be a need for human interpretation of models going forward?

AI is progressing at an incredible rate and continues to positively impact massive data processing:  https://x.com/FutureJurvetson/status/1863649174358831312 

[wxsniss]

15 hours ago, wxsniss said:

Thoughts? Anyone try this?

Part of me hopes this is just more typical hype than an actual game changer… I think we’d all miss the suspense and art of forecasting snow storms.

https://www.nytimes.com/2024/12/04/science/google-ai-weather-forecast.html

https://www.nature.com/articles/s41586-024-08252-9

“ Here we introduce GenCast, a probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, ENS, the ensemble forecast of the European Centre for Medium-Range Weather Forecasts4. GenCast is an ML weather prediction method, trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-h steps and 0.25° latitude–longitude resolution, for more than 80 surface and atmospheric variables, in 8 min. It has greater skill than ENS on 97.2% of 1,320 targets we evaluated and better predicts extreme weather, tropical cyclone tracks and wind power production.”

I actually posted the Nature paper link and portion of abstract about GenCast yesterday, lost in the Banter thread… curious about people’s thoughts.

[Great Snow 1717]

2 hours ago, Layman said:

Thank you @RDRY for posting this interesting Barron's link regarding the GenCast AI weather model:  https://www.barrons.com/news/google-says-ai-weather-model-masters-15-day-forecast-cdc5793d 

Will, or can, AI improve forecasting reliability?  What are your thoughts, opinions and/or experience?  Will there still be a need for human interpretation of models going forward?

AI is progressing at an incredible rate and continues to positively impact massive data processing:  https://x.com/FutureJurvetson/status/1863649174358831312 

It is only a matter of time before AI improves forecasting. And at some point in time will replace the need for human forecasting. I am a long-time chess player. AI has been around the chess world for a considerable period of time. AI has improved to the point that it is far superior to the world's best chess players.