Special Topic: 2026 Sustainable Global Leaders Conference & the Second Green Industry and Sustainable Consumption Expo. On October 22, a financial frontline report stated that the 2026 Sustainable Global Leaders Conference was held from October 20 to 22 at the Huangpu Expo Park in Shanghai. Professor Chen Peichang, President of the Hong Kong Quality Assurance Agency, said that artificial intelligence can accelerate ESG-related data analysis and research, but model outputs still need verification. Companies should strengthen the foundation of data governance and move ESG from compliance disclosure into business management.
Drawing on his experience participating in research with technical teams, Chen Peichang said AI can improve the efficiency of quantitative and qualitative analysis, helping complete related research faster and identify problems. However, as model capabilities increase, the scale of information and the dimensions of research are also expanding, and models can still produce errors. When dealing with long reports and complex data, even if the overall accuracy rate is relatively high, a small amount of inaccurate content is not easy to detect. Therefore, the knowledge base, rigorous attitude, and verification ability of professionals remain crucial.
He believes that work on ESG, climate resilience, Scope 3, and carbon reduction must ultimately answer whether it can bring substantive help in business development. For small and medium-sized enterprises, business leaders need to understand the reasons for investing time, money, and resources. Promoting related work should use more easily understood expressions and more convenient tools, link climate response with products and enterprise development, and give companies a clear direction.
Chen Peichang pointed out that some companies still regard ESG disclosure as compliance or public relations work, and related efforts have not entered core industries and business management. For AI to play a role, companies first need to do a good job in data-based management, complete data collection, definition, and classification, and clarify data responsibilities. The difficulties faced by large institutions often also include departmental division of labor, responsibility boundaries, and governance mechanisms.
Regarding the credibility of AI-generated content, Chen Peichang said that the current stage is still a transition period of human-machine collaboration, and human participation will still be needed in the next five to ten years. Companies should master the correct methods of using AI, carry out quality assurance and multiple screening of information sources and results; in rigorous scenarios such as scientific research, different models can be used for cross-evaluation. At the same time, companies must also pay attention to data ethics, social responsibility, information protection, and personal privacy.
He said that one of the most basic and difficult tasks in AI practice is defining and collecting data. Companies need to clarify who is responsible for data, which department manages it, and establish a corresponding responsibility mechanism. Only by making the data foundation and governance work solid can data be further transformed into competitive capabilities in products, scientific research, talent training, and process management.
Chen Peichang also said that companies should shift from awareness to action and implement the goal from ESG to intelligent management and intelligent operations.
Comments