Key facts
- China released its Artificial Intelligence Security Governance Framework 3.0 on September 14.
- The framework emphasizes AI serving the public good with development and security advancing in tandem.
- It shifts oversight from AI model outputs to AI agents throughout their life cycles.
On September 14, China's National Technical Committee 260 on Cybersecurity released the Artificial Intelligence Security Governance Framework 3.0. The document carries forward a core principle of China’s earlier Global AI Governance Initiative: AI should serve the public good, with development and security advancing in tandem. More importantly, it marks a shift in Chinese AI governance from regulating model outputs to overseeing AI agents throughout their life cycles. The framework could reset the industry’s compliance baseline, strengthen risk controls across the AI value chain and contribute a Chinese perspective to the global debate over AI governance.