Before offering generative AI services to the public, conduct required safety assessments and algorithm registration; mark all generated content

Cost: There is no monetary cost, but you must follow procedures set by internet regulators — completing them takes some time. Small personal tools that are not made available to the public do not fall under these rules.
In plain terms: If you offer generative AI services with public influence or social mobilization potential to the general public, you must perform a safety assessment and register the algorithm per national regulations before launch. Any changes or shutdowns also require filing updates or cancellation paperwork. All generated images and videos must be clearly marked. Tools meant solely for personal use and not open to the public are exempt from these requirements. The regulation itself does not specify penalties; violations are handled under the Cybersecurity Law, Data Security Law, and Personal Information Protection Law.
Benefit: Providers of generative AI services that can shape public opinion or mobilize people must conduct safety assessments, register their algorithms, and handle subsequent updates or cancellations as required by law. “Public influence” or “social mobilization potential” means the ability to affect how people think or to organize them. Providers must also add identifying marks to any generated images or videos. This regulation took effect nationwide on August 15, 2023.
Evidence grade: B
Sources:国家互联网信息办公室等七部门 (2023). 生成式人工智能服务管理暂行办法（第十二、十七条）. https://www.gov.cn/zhengce/zhengceku/202307/content_6891752.htm
Notes: The regulation itself does not state exact penalty amounts. Violations are penalized under the Cybersecurity Law, Data Security Law, and Personal Information Protection Law, which is why this item is rated B. No civil cases involving copyright or portrait rights violations caused by AI‑generated content are cited here, as the original texts were not verified on the Supreme People’s Court website.