DeepMind's WeatherNext AI model achieves breakthrough cyclone forecasting
Google DeepMind's AI model predicts hurricane tracks and intensity with unprecedented accuracy, giving forecasters an extra day of warning.
Conversation activity · last 3 days peak 2/hr
Summary, timeline and people extracted by Claude from 25 items across 5 sources · 5h ago. Quotes are verbatim.
Google DeepMind published research in Nature showing its WeatherNext AI model achieves state-of-the-art accuracy in predicting tropical cyclone tracks, intensity, and wind structure—delivering forecasting accuracy equivalent to one extra day compared to existing models. The model already supported the National Hurricane Center's forecast for Hurricane Melissa in 2025 and now generates 1,000 scenario predictions per cyclone; DeepMind is open-sourcing the model to amplify its impact across the research community.
- WeatherNext delivers an extra day of cyclone forecasting accuracy by predicting track, intensity, and wind structure simultaneously—equivalent to a decade of traditional meteorological progress.
- The model achieved real-world impact in 2025 when it helped the National Hurricane Center forecast Hurricane Melissa, enabling advance warnings; now generates 1,000 scenario predictions per cyclone.
- DeepMind is open-sourcing the WeatherNext models to amplify research community adoption, though forecasters emphasize human expertise remains critical for translating predictions into impact assessments.
- Technical community notes WeatherNext exemplifies valuable problem-specific AI applications often overshadowed by focus on large language models and demonstrates Graph Neural Networks' power in scientific prediction.
How it unfolded
-
Reaction Technical community discusses broader AI implications
Hacker News commenters note WeatherNext represents important problem-specific AI models beyond LLMs, with Graph Neural Networks being underappreciated architectures in weather forecasting.
“Everything in AI seems to be focused on LLMs lately. But in my opinion, powerful problem-specific models like this are even more interesting.”
tcumulus · Hacker News ↗ -
Reaction Weather scientists express surprise
Ars Technica and WIRED coverage highlights that DeepMind's breakthrough has surprised weather scientists, noting the model can make accurate predictions with lower-resolution weather data.
-
Analysis Forecasters emphasize human role remains critical
Weather experts clarify that while WeatherNext is a valuable tool, human judgment remains essential—impacts ultimately require expert translation, not just track and intensity forecasts.
“A hurricane is not just a track or an intensity forecast. It requires experts to translate that into what the impacts are going to be—and it's the impacts that kill people.”
Brennan (forecaster) · Bluesky ↗ -
Report Nature paper published on WeatherNext
DeepMind published research showing WeatherNext achieves state-of-the-art cyclone forecasting with three-day forecasts as accurate as prior models' two-day forecasts—equivalent to a decade of meteorological progress.
“On average, our model gives forecasters an extra day's worth of predictive accuracy: our three-day forecasts are as good as what prior models were able to provide for only the next two days.”
DeepMind · Mastodon ↗ -
Event WeatherNext models open-sourced
DeepMind announced it is open-sourcing WeatherNext 2 and WeatherNext Cyclones models used during hurricane season to empower the research community.
“By making this technology openly available, we hope to empower the research community and amplify AI's impact in building more resilient communities.”
DeepMind · Mastodon ↗ - 83 weeks quiet
-
Event Hurricane Melissa forecast breakthrough
WeatherNext model helped the National Hurricane Center predict Hurricane Melissa's rapid intensification and Jamaica landfall, enabling advance warnings during the 2025 hurricane season.
What people are saying verbatim
“Our three-day forecasts are as good as what prior models were able to provide for only the next two days. This scale of improvement corresponds roughly to a decade's worth of meteorological progress.”
DeepMind, Research organization · DeepMind official blog ↗ · Aug 5
“WeatherNext enables accurate cyclone forecasts that can give an extra day of warning. Now we are open sourcing the model.”
DeepMind, Research organization · Mastodon ↗ · Aug 5
“It's a black box at the end of the day, but that gives physicists a signal that something is happening that was not previously understood.”
Alet, DeepMind researcher · Ars Technica ↗ · Aug 7
“A hurricane is not just a track or an intensity forecast. It requires experts to translate that into what the impacts are going to be—and it's the impacts that kill people.”
Brennan, Weather forecaster · Ars Technica ↗ · Aug 7
“Everything in AI seems to be focused on LLMs lately. But in my opinion, powerful problem-specific models like this are even more interesting.”
tcumulus, Hacker News commenter · Hacker News ↗ · Aug 7
“I'm very excited about scientific discovery. I think AI is giving us new tools to poke into the laws of the universe.”
Alet, DeepMind researcher · Ars Technica ↗ · Aug 7
“A reminder that AI is more than LLMs and that other prediction models can do useful stuff”
garvanwalshe.org, Bluesky commenter · Bluesky ↗ · Aug 8
Voices from the web unedited
-
“In a paper published on Thursday in Nature, researchers show that the WeatherNext AI model can predict cyclones with unprecedented accuracy. On average, it gives forecasters a day more lead time than existing models;”
-
I just discovered typhoon/cyclone predictions and they're insane. I get mine via https://zoom.earth (whose iPhone app is terrific).Here's a selection from Typhoon Dolphin, currently sitting off the east coast of China. Dolphin continues its slow, trochoidal Z motion, generally heading westward deeper into the East China Sea. Over the past 12…
-
🎉 WeatherNext：Google DeepMind 的 AI 全球中短期天气预报模型 来自 http:// blog.xlap.top/post/tech/2026-0 8-10/weathernext/ # 中文独立博客推送
-
A reminder that AI is more than LLMs and that other prediction models can do useful stuff
-
> One key limitation of our approach is in how uncertainty is handled. We focused on deterministic forecasts and compared against HRES, but the other pillar of ECMWF’s IFS, the ensemble forecasting system, ENS, is especially important for 10+ day forecasts. The non-linearity of weather dynamics means there is increasing uncertainty at longer lead…
-
Predicting big weather events is not that hard even with 50 year old technology.What's hard is predicting details, like exactly where it will rain, what the slope of the beach is today (many people don't even know this changes drastically daily and why it is important), wave height, ocean depth today where people swim, water temperature…
-
Everything in AI seems to be focused on LLMs lately. But in my opinion, powerful problem-specific models like this are even more interesting. The SOTA AI models used in weather forecasting are already outperforming the classic NWP models while being orders of magnitude more efficient (inference). Most are based on multi scale (hierarchical) Graph…
-
I couldn't find a client of any sort for WeatherNext, so I had Claude write a simple one:
-
Maybe was this that was the last drop for Sundar.Demis: "I have a new amazing breakthrough"Sundar: "Great! We really need a answer to Sol and Fable"Demis: "They are completely owned in typhoon forecasting"
-
From the tagline in the article: "WeatherNext enables accurate cyclone forecasts that can give an extra day of warning. Now we are open sourcing the model."