China's Cybersecurity Standardization Committee Releases AI Safety Governance Framework 3.0
China's National Cybersecurity Standardization Technical Committee has released version 3.0 of its framework for governing the safety of artificial intelligence. The new framework addresses emerging trends and challenges in AI development.
China's National Cybersecurity Standardization Technical Committee (网安标委) released version 3.0 of its AI Safety Governance Framework on Thursday, September 14. The announcement was made during the opening ceremony of the National Cybersecurity Week.
The new framework, titled 《人工智能安全治理框架 3.0》, was developed under the guidance of the Cyberspace Administration of China (CAC, 网信办). It aims to address new security challenges and risks posed by the rapid advancement of artificial intelligence.
Framework 3.0 builds upon previous versions, 1.0 and 2.0, released in 2024 and 2025. It adheres to principles emphasizing a human-centered approach and ensuring security and controllability. The framework updates risk classifications and optimizes technical and governance measures to manage AI-related risks.
The goal of the framework is to foster a shared understanding of AI safety governance and enhance risk prevention and response capabilities. This aims to ensure that AI technology benefits humanity safely and responsibly.