China Launches Crackdown on AI Misuse

China's Cyberspace Administration of China (CAC), together with the Ministry of Public Security and others, has launched the 2026 edition of the months-long annual Qinglang (清朗) special campaign. The focus is on AI-based fraud, deepfake impersonation, disinformation, and violations of personal information and intellectual property.

• In the same campaign in 2025, over 3,500 AI-related products were taken down, and about 960,000 pieces of illegal or harmful content and roughly 3,700 accounts were processed.
• China implemented the AIGC labeling mandate, requiring visual and technical labels on all AI-generated text, images, audio, and video, on September 1, 2025, and has broadened its enforcement toolkit with overlapping measures including LLM pre-filing and security review, Deep Synthesis regulations, and the Interim Measures for Humanoid AI Interaction Services, set to take effect on July 15, 2026.
• The CAC has identified as key enforcement targets: fraud impersonating relatives or acquaintances, unauthorized recreation of deceased persons' faces or voices, mass fake accounts and organized public opinion manipulation, unregistered generative AI services, and use of training data that infringes on rights; as of March 2025, 346 generative AI services had completed filing.
• Violations can result in administrative penalties, service suspension, or time-limited corrective orders, with serious cases potentially referred to public security authorities. The CAC, MIIT, and the Ministry of Public Security have also jointly issued a 2026 personal information protection enforcement agenda targeting seven sectors, including internet advertising, education, healthcare, and illegal data trading.

The timing also overlaps with the White House's April 23 accusation of 'industrial-scale' AI theft by China, but this crackdown has been characterized as a separate enforcement effort focused on protecting rights holders and users within China.