The Frontier AI and Agent Safety Forum was held in Shanghai on Sunday morning, July 19, as part of the 2026 World Artificial Intelligence Conference (WAIC), bringing together leading experts to explore new approaches to artificial intelligence safety and global governance.
More than 20 renowned AI scientists, researchers and policymakers from around the world attended the forum, where they jointly unveiled several significant global research reports and institutional achievements in the field of AI safety. Participants also engaged in in-depth discussions on frontier AI risk management and the future governance of the global AI industry.
The forum focused on three key areas: frontier AI risk monitoring, risks associated with uncontrolled AI agents and emergency response mechanisms, and global perspectives on frontier AI risk management.
Throughout the day, experts delivered keynote speeches and participated in roundtable discussions addressing the challenges and opportunities surrounding the safe development of advanced AI systems.
A major highlight of the event was the joint release of five key research outcomes. Covering advanced technological tools, AI classification frameworks and evaluation standards, emergency response strategies, and comprehensive assessments of the future AI industry landscape, these achievements aim to contribute to the development of a systematic and internationally coordinated framework for AI safety governance.
Source: Xuewei, Touhid and Jinnia, China Media Group (CMG).