Climate
The part Chinese AI could play in predicting extreme weather
Publication: Dialogue Earth
Author: Na Xu
Source published: 30/09/2026, 13:31:52 UTC
Briefing created: 02/10/2026, 10:02:22 UTC
Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 02/10/2026, 09:13:24 UTC.
Briefings
Chinese AI models achieved significant breakthroughs in weather forecasting, with the Shanghai Artificial Intelligence Laboratory's Fengwu model predicting Typhoon Doksuri's path 24 hours in advance with minimal error. However, these models struggle to accurately predict extreme events like record-breaking rainfall or typhoon intensity due to data limitations. The China Meteorological Administration launched the Mazu system in July 2025, combining satellite data with AI, but experts caution that without comprehensive data coverage, AI may fail to forecast rare or non-existent weather patterns.
Key points
- The Shanghai Artificial Intelligence Laboratory's Fengwu model predicted the path of Typhoon Doksuri 24 hours in advance with an error of only 38.7 km, marking a major milestone in Chinese AI weather forecasting.
- AI models currently fail to predict extreme weather events like record-breaking rainfall or typhoon intensity because they lack sufficient data in their training sets, leading to systematic underestimation of frequency and intensity.
Context and limitations
AI models require extensive historical data to function effectively; without it, they cannot predict rare or non-existent weather events. Current systems like Fengwu and Mazu show promise but remain limited in forecasting extreme convective weather and localised flash floods due to data gaps.
Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 02/10/2026, 10:02:22 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.
View archived briefing on IPFS
bafybeieurwhyv7mes7d7ucgcbuv5vtkairg7z5vbzrjaxduffcgrmw4wim