Recently, the Central Cyberspace Administration of China launched the "Clear and Bright - Special Action to Rectify the Abuse of AI Technology". Since April 2025, a series of rectification efforts have been carried out targeting the abuse of AI technology. The action focused on the infringement of public rights caused by AI face swapping and voice simulation technologies, as well as the misleading of the public due to the lack of content labeling. After the first phase of efforts, cyberspace administration departments at all levels joined forces and achieved significant results.

Artificial Intelligence Mechanical Arm AI (2)

Image source note: Image generated by AI, image authorization service provider Midjourney

During the rectification process, over 3500 non-compliant mini-programs, applications, and intelligent entities were dealt with, more than 960,000 pieces of illegal information were cleared, and over 3700 non-compliant accounts were handled. These measures aim to cut off the marketing and traffic diversion channels for non-compliant products, urge major websites and platforms to strengthen technical security guarantees, and accelerate the implementation of content labeling for generated synthetic content.

Cyberspace administration offices at all levels took various measures to actively promote the rectification work. For example, the Beijing Cyberspace Administration opened up reporting channels and formed a working model of "user marking - platform verification - joint disposal", successfully handling 26 reporting cases. The Shanghai Cyberspace Administration organized regulatory promotion activities covering more than 400 enterprises to ensure the launch of explicit labeling norms. The Zhejiang Cyberspace Administration strengthened supervision of AI applications and website platforms, intercepting and clearing more than 25.5 million illegal pieces of information in total.

Major platforms are also actively fulfilling their responsibilities. Tencent Company standardized application management, optimized patrol mechanisms, and handled more than 570 non-compliant products. Weibo cleaned up more than 4800 pieces of non-compliant content through multiple review channels. Douyin established a "red-blue confrontation" mechanism, fixed potential security vulnerabilities, and enhanced the ability to identify false information.

Key points:

🌐  The "Clear and Bright - Special Action to Rectify the Abuse of AI Technology" was launched in April 2025 and handled more than 3700 non-compliant accounts.  

📊  In the first phase of rectification, more than 3500 non-compliant AI products were handled, and over 960,000 pieces of illegal information were cleared.  

🔍  Cyberspace administration offices at all levels actively took measures to promote the implementation of explicit labeling standards and platform responsibilities, ensuring network security.