Key Takeaways Alibaba unveiled Baguan-Seasonal at COP30 in Belem, Brazil, capable of forecasting climate conditions up to 12 months ahead. The new model introduces probabilistic forecasting to address...
Alibaba has introduced Baguan-Seasonal, a new artificial intelligence model designed for long-range climate forecasting, at the United Nations Climate Change Conference (COP30) in Belem, Brazil.
The model extends prediction capabilities from several months to a full year ahead, marking a significant advancement in AI-powered climate science.
The Baguan-Seasonal model builds on its predecessor, Baguan, released in November 2024 by Alibaba's DAMO Academy.
The original model delivers short-term weather forecasts from one hour to ten days in advance with hourly updates and spatial resolution down to one-kilometer grids.
The evolution from short-term weather prediction to long-range climate forecasting addresses a persistent challenge in the field.
Reliably modeling climate uncertainty has proven difficult due to the Earth's inherent complexity and dynamic systems. Most existing AI models struggle to capture probabilistic nuances essential for accurate long-range predictions.
Baguan-Seasonal tackles this challenge through a novel tokenization strategy and a mixed-scale conditioning mechanism designed to handle high-dimensional climate data across spatial and temporal scales.
The model enables AI to understand complex interactions between the Earth's atmosphere and oceans, which is critical for extending the prediction horizon to 12 months.
The technology has already demonstrated real-world value in disaster preparedness. Alibaba has deployed the Baguan system across multiple applications in China over the past year.
In collaboration with the Zhejiang Meteorological Observatory, a customized model was developed to predict tropical cyclone paths and intensities in 2025.
The model proved its worth during Tropical Cyclone Co-May, which struck eastern China in late July 2025.
According to the Zhejiang Meteorological Observatory, the Baguan-based system achieved 50 percent higher accuracy in predicting the cyclone's intensity compared to other AI models.
This precision aided the evacuation of approximately 97,000 citizens from coastal and low-lying areas in Zhejiang province.
Alongside Baguan-Seasonal, Alibaba introduced Baguan-S2S, an AI model designed for sub-seasonal weather prediction covering 14 to 42 days ahead.
Research found that Baguan-S2S could capture early signals of the North Atlantic Oscillation, a key atmospheric pattern influencing weather across the North Atlantic, Europe, and North America, four weeks in advance.
The model predicted a pronounced cold anomaly of approximately six degrees Celsius in Europe, closely matching actual observations and extending the forecast lead time by approximately one week compared to the European Centre for Medium-Range Weather Forecasts.
The announcement coincided with the release of Alibaba's AI for Good Report 2025, which debuted at COP30.
The report details the company's broader AI applications for environmental and social benefit, including green datacenter operations, healthcare innovations, and educational programs.
Alibaba Cloud's green computing solutions enabled enterprise customers to avoid 11.19 million metric tons of CO2 emissions in fiscal year 2025.
The company has implemented a comprehensive green strategy encompassing green energy, products, architecture, operations, and services.
These initiatives include computing-power synergy, intelligent job scheduling, AI-enhanced digital twin systems, and advanced cooling and power infrastructure.
In the energy sector, Baguan models are now operational in several Chinese cities, including locations in Shandong, Zhejiang, and Beijing, to enhance renewable energy forecasting accuracy.
The system supports grid planning by providing reliable predictions under complex and volatile weather conditions.
The announcement comes as AI emerges as a prominent topic at COP30, with both opportunities and concerns under discussion.
Proponents argue that AI could help reduce global emissions by three to five billion tons over the next decade through improvements in agriculture, transport, energy management, weather forecasting, and disaster risk assessment.
However, critics have raised concerns about the environmental footprint of AI systems themselves, including electricity demand, water consumption, and emissions from data centers.
Key takeaways OpenAI has signed a multi-year, $10 billion agreement with AI chipmaker Cerebras Systems to secure computing infrastructure. The deal will deliver 750 megawatts of computing power throug...
KEY TAKEAWAYS: xAI implemented restrictions preventing Grok from editing images of real people in revealing clothing after global backlash California Attorney General Rob Bonta launched investigation...
Key takeaways TSMC reported fourth-quarter net income of NT$505.74 billion, beating analyst expectations and setting a new record. Fourth-quarter revenue reached NT$1.046 trillion, up 20.5% year over...
