China Pursues AI Legislation to Ensure Systems Remain Under Human Control

China is accelerating its research into comprehensive AI legislation designed to prevent artificial intelligence from evading human control. The move signals a strategic shift toward formalizing systemic safeguards as the technology evolves toward greater autonomy.

This effort was underscored by Sun Xiaobo, the Coordinator for AI Affairs of the Ministry of Foreign Affairs. During a convening on AI and Human Development, which was attended by UN Deputy Secretary General Amina Mohammed, Sun emphasized that AI should "always remain under human control." The push for broader legislation reflects a growing urgency in Beijing to ensure that AI systems remain secure and controllable, preventing scenarios where autonomous systems might diverge from human intent.

The Race for Superintelligence

This regulatory drive occurs against the backdrop of an intense global competition for AI supremacy, primarily between the United States and China. Both nations are currently attempting to balance the pursuit of rapid technological breakthroughs with the management of existential risks. Central to this tension is the prospect of "superintelligence"—the development of autonomous systems capable of bypassing human overrides or operating outside the parameters set by their creators.

Global Regulatory Implications

If China formalizes legislation specifically targeting the risk of AI "escape," it could establish a significant global precedent for AI safety standards. Such a move would shift the conversation from voluntary ethical guidelines to binding regulatory frameworks. This transition could potentially catalyze the creation of international treaties focused on AI containment, forcing other major tech powers to align their domestic laws with a standardized set of safety protocols to avoid a "race to the bottom" in safety oversight.

The Path Forward

While the intent to maintain human supervision is clear, the specific mechanisms of the proposed legislation remain under research. Observers are now watching to see whether China will implement technical "kill switches" or mandatory auditing requirements for frontier models. It remains to be seen how these domestic laws will integrate with existing international governance efforts or if they will create a fragmented regulatory landscape between East and West. The outcome of this research will likely determine whether the global community moves toward a unified safety standard or a bifurcated system of control.