China Develops AI Safety Rules to Prevent Loss of Human Control

China is actively developing regulatory frameworks and safety guidelines to address the risk of advanced artificial intelligence systems escaping human control, even as it competes with the United States in the global AI race. Chinese authorities have explicitly included scenarios where future AI could autonomously obtain resources, replicate itself, and seek power, creating a risk of competing with humans for control. China's national AI safety framework, first released in September 2024 and expanded in September 2025, outlines these loss-of-control risks. The expanded version describes the possibility of a sudden and unexpectedly large leap in intelligence before an AI system acquires resources and seeks power. It also introduced a governance principle of trusted application, preventing loss of control. In May, Chinese regulators issued joint guidelines for AI agents, which can perform tasks more independently than traditional chatbots. The rules require developers to improve their ability to detect, intervene in, and recover from harmful behavior.