WIC Specialized Committee on AI holds symposium on coordinated governance of frontier AI safety risks

The World Internet Conference (WIC) Specialized Committee on Artificial Intelligence held the Symposium on Coordinated Governance of Frontier AI Safety Risks in Beijing on July 15.

Ren Xianliang, secretary-general of the WIC, delivered the opening remarks. Amandeep Singh Gill, United Nations under-secretary-general and special envoy for digital and emerging technologies, addressed the symposium via video.

Participants noted that rapid advances in frontier AI, including large models and AI agents, are reshaping economic and social development and accelerating global technological innovation. 

At the same time, risks related to model misuse, cybersecurity, disinformation and cross-border diffusion are becoming increasingly interconnected — making frontier AI safety risks more complex, cross-sectoral and unpredictable. These challenges called for stronger capabilities in risk identification, assessment, monitoring, early warning, intervention and prevention, they said.

Participants agreed that no single country, institution or technical approach can effectively address these emerging challenges alone. Closer collaboration among government agencies, international organizations, businesses, universities, research institutions and think tanks is essential to building a more open, inclusive and effective governance framework for frontier AI safety.

They put forward three recommendations for strengthening coordinated governance of frontier AI safety risks.

They called for enhancing risk identification and assessment. Governance efforts should continuously improve capabilities in identifying, evaluating, monitoring and assessing emerging risks associated with evolving AI capabilities and wider deployment, while strengthening forward-looking analysis of major, long-term and cross-sector risks.

More efforts are needed to align governance frameworks and rules, they said, to promote coordination among government regulation, corporate governance, industry self-regulation and public oversight. This will enhance the interoperability of technical standards, evaluation methodologies, governance procedures and accountability mechanisms and support AI innovation consistent with scientific ethics, the public interest and safety requirements.

They also highlighted the need to expand pathways for international cooperation. The WIC need to give full play to its role as an international platform for exchanges and cooperation, enhancing dialogue on key issues in frontier AI safety governance and fostering consensus and cooperation on risk understanding, governance mechanisms, capacity building and coordinated responses.

At the keynote speech session, prominent guests talked about interoperability in global artificial intelligence governance, model safety assessment, agent safety, AI-native security systems, trusted platform governance, and international collaboration.

Speakers included John Higgins CBE, chairman of the International AI Governance Association; Zeng Yi, Wu Yuzhang chair professor of the Gaoling School of Artificial Intelligence at Renmin University of China and dean of the Beijing Institute of AI Safety and Governance; Pan Jianfeng, chief scientist and senior vice-president of 360 Group; Wei Kai, director of the Artificial Intelligence Research Institute at the China Academy of Information and Communications Technology; and Wang Wei, COO of Weibo, CEO of Sina Mobile and head of the Sina AI Media Research Institute. Seán Ó hÉigeartaigh, research professor at the Centre for the Future of Intelligence of University of Cambridge, delivered a video speech.

The roundtable dialogue, moderated by Zeng, focused on "Coordinated Response to Frontier AI Risks." 

Participants exchanged views on topics including the risks posed by emerging technologies such as intelligent agents, the interoperability of governance mechanisms, enterprise practices, and international collaboration.

Those at the roundtable discussion were Jimena Sofia Viveros Alvarez, member of the UN High-Level Advisory Body on AI and managing director and CEO of IQuilibriumAI; Cyrus Hodes, member of the OECD Expert Group on AI Futures and founder of AI Safety Connect; Lin Mingfeng, vice-president of DAS-Security; and Liu Jiaqi, chief expert on AI Governance Industry Development at Huawei.

The symposium was organized by the Beijing Institute of AI Safety and Governance and co-organized by 360 Group. 

More than 120 people, including members of the WIC Specialized Committee on AI, WIC members and representatives from government agencies, foreign diplomatic missions in China, international organizations, research institutions and enterprises, attended the event on-site and online.