en.Wedoany.com Reported - Recently, the National Standardization Administration officially issued the mandatory national standard plan for "Basic Requirements for Agent Application Security" (Plan No.: 20263116-Q-252). This standard is the world's first mandatory national standard targeting agent security, filling an international gap in mandatory standards for agent security in the public service sector. It represents a strategic initiative by China to improve the comprehensive governance system of artificial intelligence. The standard is managed by the Cyberspace Administration of China and drafted by China Mobile Communications Group Co., Ltd.
Currently, the industrialization of agents is accelerating comprehensively, with technological iteration and industry penetration exceeding expectations. At the same time, the underlying security logic of agents has undergone fundamental changes—security risks have expanded from traditional content output to autonomous action levels, with issues such as information leakage, permission loss, tool abuse, and intent deviation increasingly posing real threats. Major global economies have intensively introduced agent governance rules, and the competition over technology and standards has fully escalated. Against the dual backdrop of industry and governance, establishing unified mandatory security standards is timely.
This standard focuses on agent products and services for the public, closely centering on preventing key security risks and safeguarding security bottom lines to build a security compliance framework. It also emphasizes transforming macro security governance red lines into implementable and testable technical compliance requirements, providing a legal foundation for subsequent industry supervision, market access, and technical evaluation, while reserving reasonable development space for technological innovation and scenario implementation.
Based on long-term technical accumulation in communication networks, computing infrastructure, and artificial intelligence, and leveraging preliminary research in multiple domestic and international standards organizations such as the ITU Telecommunication Standardization Sector, the National Cybersecurity Standardization Technical Committee, the Artificial Intelligence Standardization Technical Committee of the Ministry of Industry and Information Technology, and the China Communications Standards Association, China Mobile has built a security technology system covering the entire lifecycle of agent tasks. This successful approval marks China Mobile's AI security governance technical capabilities entering the domestic leading tier.
During the critical period of global agent application and industrial strategic competition, rule discourse power has become a core competitive advantage. As a forward-looking governance tool for coordinating development and security, this mandatory national standard plan not only provides a benchmark for preventing systemic risks in the agent industry and safeguarding network and data security bottom lines but also lays the foundation for building a secure, reliable, and controllable AI industry ecosystem. At the same time, it represents an important practice for China to deeply participate in global AI governance, export Chinese governance ideas, and strive for discourse power in international rules in the agent field.
