【Speaker】：Dr. Gang Huang (Chinese Academy of Sciences)

【Title】：Building an Open Community for a Super Artificial Intelligence Earth System Driven by Both Data and Physics

【Time】：3:00pm, Jan 26, 2026.

【Location】：Beijing Lab for System Engineering of Carbon Neutrality

【About Report】：AI has been deeply integrated into basic disciplines such as life science, physics and chemistry. China and the United States have formed a "dual-core driving" pattern. China performs prominently at the application layer such as material design and drug development, while the United States maintains a leading position in basic algorithms such as AlphaFold and DeepMind. As a representative of interdisciplinary science, meteorological science has long been the focus of industry-university-research attention. Artificial intelligence is currently promoting the transformation of meteorological science from "experience-driven" to "data-physics dual-driven". The accuracy of China's AI weather forecast in 2024 increased by 127% compared with that in 2018, yet bottlenecks including physical interpretability and extreme event prediction still need to be broken through. In the future, it is necessary to focus on developing hybrid modeling, digital twins and global collaboration systems to achieve leapfrog upgrading from precise forecasting to disaster chain risk management. This report proposes building an open-source, collaborative and inclusive platform that integrates physical models such as climate dynamics and ecological evolution with AI technologies such as deep learning and reinforcement learning. Meanwhile, ideas of super AI and modular development are put forward to promote high-precision simulation and prediction of earth system science and tackle global issues including climate change and resource management.