Alibaba’s DAMO Academy says AI can forecast major weather shocks a year ahead at COP30

A bold COP30 pitch

Alibaba (阿里巴巴) research arm DAMO Academy (达摩院) used the 30th UN Climate Change Conference (COP30) to showcase an upgraded AI weather system it says can predict major climate-linked events up to 12 months in advance. The model—dubbed “Baguan” (八观)—targets phenomena such as El Niño and severe cold waves. Can AI really see that far ahead? The claim, if validated, would stretch typical forecast lead times and strengthen long‑range preparedness for governments and industry.

Inside “Baguan”

According to the institute, Baguan has already been deployed in parts of China, including Zhejiang, Shandong, and Beijing, where it reportedly cut typhoon intensity forecast errors by more than 50%. The team said the latest upgrade focuses on long‑term climate warning, with year‑ahead outlooks for globally significant events. Independent benchmarking and peer‑reviewed validation were not disclosed; it has been reported that the system draws on large‑scale multi-source data and foundation‑model techniques to improve spatiotemporal prediction.

Why it matters

For Western readers unfamiliar with DAMO Academy, it is Alibaba’s central R&D hub, known for pushing applied AI across commerce, logistics, and science. If Baguan’s performance holds up, the implications are broad: earlier disaster preparedness, grid and commodity planning, insurance risk modeling, and more targeted evacuations in typhoon‑prone regions. China, which faces frequent landfalling typhoons, has invested heavily in numerical weather prediction and now increasingly in AI-augmented forecasting.

The geopolitical backdrop

The announcement lands amid intensifying competition in AI and ongoing U.S. export controls on advanced chips, which have constrained Chinese labs’ access to top-tier training hardware. Climate and weather data, however, still flow through multilateral channels like the World Meteorological Organization—an arena where scientific cooperation often persists despite geopolitical strain. The key test now: rigorous, third‑party evaluation against benchmarks from agencies such as ECMWF and NOAA to determine whether year‑ahead skill truly exceeds today’s best-in-class models.