On September 13, Central Broadcasting Network reported from Shanghai (Reporter Yin Yuting) that on September 12, the 2025 Inclusion·Bund Conference held an insightful forum themed “AI for Good: Safe Development – Ethics and Governance of Large Models in Rigorous Industry Applications,” co-hosted by the Xinhai Research Institute and the Ant Group Technology Ethics Committee. Professor Jiang Xiaojun from the Chinese Academy of Social Sciences University, and former Deputy Secretary-General of the State Council, attended the forum and delivered a keynote speech titled “AI Governance: How to Move from Consensus to Action.”
Professor Jiang Xiaojun from the Chinese Academy of Social Sciences University and former Deputy Secretary-General of the State Council delivered a speech. Photo provided by the interviewee
Over 20 authoritative experts from academia, industry, and regulatory bodies engaged in in-depth discussions on the ethics and governance of artificial intelligence, particularly the risks and guidelines for the application of large models in rigorous industries such as healthcare, aiming to promote the coordinated development of technological innovation and ethical standards.
Huang Chengqing, Vice Chairman of the China Internet Association and a member of the Xinhai Research Institute, delivered a speech. Photo provided by the interviewee
“Especially in the healthcare sector, AI’s decisions are directly related to life, health, and social equity. Issues such as data misuse, algorithmic bias, and ambiguous responsibility urgently need to be addressed,” stated Huang Chengqing during the forum. He emphasized that building a trustworthy, reliable, and traceable AI governance system has become an inherent requirement for technological development and an inevitable choice for social responsibility.
“Artificial intelligence provides rich perspectives and value for the development of various fields, but it is also challenging the primary position of humanity,” pointed out Zheng Qinghua, an academician of the Chinese Academy of Engineering and Secretary of the Party Committee at Tongji University. He stated that aligning AI with human values is fundamental to ensuring that AI technology benefits humanity. Academician Zheng believes that large models should learn human values and preferences through reinforcement learning to align AI with human values, creating safe and ethical AI systems while managing long-term risks in technological development. Meanwhile, humans must also align with AI, focusing on ensuring that people responsibly apply AI.
“Global AI governance is moving towards promoting development and flexible regulation, establishing an agile governance system to encourage proactive self-discipline in the industry through policy guidance, and building a responsible AI safety system with an inclusive and cautious attitude,” explained Wei Liang, Vice President of the China Academy of Information and Communications Technology. He mentioned that as of July 2025, the “AI Safety Commitment” has been disclosed by 18 leading industry companies, including Alibaba, Baidu, and Ant Group, forming 20 practical labels and 43 safety cases. This not only pushes China’s AI safety governance into a new phase of systematization, transparency, and practicality but also contributes a “human-centered, AI for good” solution from China to the world.
Zhou Zhifeng, Senior Vice President of Ant Group and Co-Chair of the Technology Ethics Committee, delivered a speech. Photo provided by the interviewee
“From a risk resilience perspective, Ant believes that merely isolating and preventing risks is not enough. We hope to enhance our capabilities through dynamic mechanisms such as early warning, testing, and remediation,” said Zhou Zhifeng, who introduced Ant’s progress and ethical considerations in the field of AI large models as a representative of the business sector.
“Taking AI medical application AQ as an example, the Q&A provided to users has an inspection mechanism. Health model scenarios are examined from five major aspects: medical ethics, medical safety, medical compliance, general safety, and medical professionalism,” Zhou Zhifeng stated. He emphasized that Ant Group is not only laying a solid foundation for AI ethics through long-term technical investment but has also formulated fundamental technology ethics principles of “equality, respect, trust, and responsibility” around artificial intelligence; by establishing a Technology Ethics Committee, they facilitate long-term effective interaction with internal departments, government guidelines, industry experts, and user feedback.
“When AI is widely used in healthcare, to what extent does human free will remain? How should individuals maintain their dignity? This seemingly is a pain point in the industry, but it actually involves fundamental issues regarding human survival and development,” said Professor Xiao Yanghua from Fudan University, who hosted the roundtable discussion segment of the forum, further focusing on “AI Large Models + Healthcare.” Experts in the fields of healthcare and ethics from Fudan University, Central South University, West China Hospital, and Shanghai Jiao Tong University engaged in intense discussions on the ethical boundaries of AI diagnosis and treatment, division of responsibilities, and regulatory frameworks.
“I believe that the greatest contribution of developing large medical models to society is the potential to promote healthcare equity and justice,” stated Professor Wang Guoyu from the Department of Philosophy at Fudan University. He noted that addressing the uneven distribution of healthcare resources in different regions of China, AI applications in healthcare not only serve as good assistants for doctors in remote areas but also bring hope and health blessings to patient groups that struggle to access and cover superior medical resources. However, he added, “Beyond efficiency, there are many difficult-to-quantify aspects, such as patients’ emotions, psychological needs, and mental states.” Wang emphasized that healthcare providers should not only treat patients’ ailments but also consider patients as whole individuals and take their life experiences into account.
“How can patients and doctors achieve a consensus on health matters? Many wearable devices can assist with this. For instance, heart rate, blood pressure, and daily exercise steps can be monitored, and reminders for medication can also be provided through wearable devices,” reflected Zhao Yan, Director of the Information Center at Ruijin Hospital affiliated with Shanghai Jiao Tong University School of Medicine. He noted that the emergence of AI enables patients to gain a more scientific understanding of their conditions before diagnosis, and promotes adherence to doctors’ treatment plans during and after diagnosis, thereby enhancing treatment outcomes.
“The exploration of AI governance never stops, and I believe today’s discussion is an important step towards perfecting AI governance rules,” Huang Chengqing stated. He mentioned that the Xinhai Research Institute will continue to collaborate with the Ant Group Technology Ethics Committee to promote cooperation between industry, academia, and research, contributing to the construction of a safe and trustworthy AI ecosystem, allowing technology to truly benefit humanity.è¿åæçï¼æ¥çæ´å¤
