Promulgated jointly by the Cyberspace Administration of China, the National Development and Reform Commission, the Ministry of Education, the Ministry of Science and Technology, the Ministry of Industry and Information Technology, the Ministry of Public Security, and the National Radio and Television Administration on July 10, 2023
Effective: August 15, 2023
Table of Contents
Interim Measures for the Administration of Generative Artificial Intelligence Services
Article 1 — These Measures are formulated in accordance with the Cybersecurity Law of the People’s Republic of China, the Data Security Law of the People’s Republic of China, the Personal Information Protection Law of the People’s Republic of China, the Law of the People’s Republic of China on Science and Technology Progress, and other relevant laws and administrative regulations for the purpose of promoting the sound development and standardized application of generative artificial intelligence, safeguarding national security and public interests, and protecting the lawful rights and interests of citizens, legal persons, and other organizations.
Article 2 — These Measures shall apply to the research, development, and provision of generative artificial intelligence services to the public within the territory of the People’s Republic of China (hereinafter referred to as “generative AI services”). For the purposes of these Measures, “generative artificial intelligence” refers to technology that generates content such as text, images, audio, and video based on algorithms, models, and rules. Where the State has other provisions on the use of generative AI technology to engage in activities such as news publishing, film and television production, and artistic creation, such provisions shall prevail. Where generative AI services provided by industry organizations, enterprises, education and scientific research institutions, public cultural institutions, and relevant professional institutions are not provided to the public, these Measures shall not apply.
Article 3 — The State shall adhere to the principle of placing equal emphasis on development and security, combining promotion and regulation with both innovation-driven development and governance in accordance with the law, and adopting effective measures to encourage the innovative development of generative AI technology. The State shall implement inclusive, prudent, and classified and graded supervision of generative AI services.
Article 4 — The provision and use of generative AI services shall comply with the requirements of laws and administrative regulations, respect social morality and ethics, and shall not generate content that incites subversion of State power or the overthrow of the socialist system, endangers national security and interests, damages the image of the State, incites secession or undermines national unity or social stability, or promotes terrorism, extremism, ethnic hatred, or ethnic discrimination, or is violent, obscene, or pornographic, or is false information, or may disrupt economic or social order.
Article 5 — When providing and using generative AI services, the following provisions shall be observed: (1) adherence to the socialist core values, and shall not generate content prohibited by laws and administrative regulations; (2) measures shall be taken during algorithm design, training data selection, model generation and optimization, and service provision to prevent discrimination based on ethnicity, religion, nationality, region, gender, age, occupation, health status, and other grounds; (3) respect for intellectual property rights and business ethics, use of data and foundational models from lawful sources, and shall not use algorithms, data, platforms, or other advantages to engage in monopolistic or unfair competition practices; (4) respect for the lawful rights and interests of others, shall not endanger the physical or mental health of others, and shall not infringe upon the right of portrait, reputation, honor, privacy, or personal information of others; (5) based on the characteristics of the service type, effective measures shall be taken to improve the transparency of generative AI services and enhance the accuracy and reliability of the generated content.
Article 6 — The State shall encourage innovation and application of generative AI technology in various industries and fields, support collaboration among industry organizations, enterprises, education and scientific research institutions, public cultural institutions, and relevant professional institutions in areas such as technological innovation, data resource construction, transformation of technological achievements, and risk prevention in generative AI.
Article 7 — Providers of generative AI services (hereinafter referred to as “providers”) shall carry out training data processing activities such as pre-training and optimization training in accordance with the law, abide by the following provisions: (1) use data and foundational models from lawful sources; (2) where personal information is involved, obtain the consent of the individual or comply with other circumstances provided by laws and administrative regulations; (3) take effective measures to improve the quality of training data and enhance the authenticity, accuracy, objectivity, and diversity of training data; (4) comply with the provisions of laws and administrative regulations and the relevant requirements of the State on the management of training data such as network information security and personal information protection.
Article 8 — Where generative AI services are provided to the public, the provider shall conduct a safety assessment and perform algorithm filing, modification, and cancellation filing procedures in accordance with the provisions on the administration of algorithmic recommendations for Internet information services. Where the generative AI technology used has the attribute of public opinion or the ability of social mobilization, the provider shall also undergo a security assessment in accordance with the relevant State regulations.
Article 9 — Providers shall assume the responsibility of information producers of the generated content and the responsibility of personal information processors of personal information. Providers shall sign service agreements with users of generative AI services to clarify the rights and obligations of both parties. Providers shall clearly and publicly disclose the applicable user groups, occasions, and uses of the service, and take appropriate measures to prevent user groups from becoming overly dependent on or addicted to the generated content.
Article 10 — Providers shall guide users in scientifically understanding and rationally using the content generated by generative AI, and shall not use the generated content to disseminate false or harmful information, engage in cyber fraud, incite social conflicts, or disrupt social order. Providers shall, as necessary, label the generated content such as images and videos in accordance with the relevant State regulations.
Article 11 — Providers shall establish a sound complaint and reporting mechanism, set up convenient channels for complaints and reports, promptly accept and handle public complaints and reports, and provide feedback on the results of handling. Where a provider discovers or learns that its generative AI service generates content prohibited by laws and administrative regulations, it shall promptly take measures such as stopping generation, stopping transmission, and eliminating the content, and take measures such as model optimization training to make rectifications, and shall report to the relevant competent department.
Article 12 — Providers shall set up labels for content generated by generative AI that may cause confusion or misidentification among the public in accordance with the Measures on the Administration of Deep Synthesis of Internet Information Services. Where a provider discovers that a user uses generative AI services to engage in illegal activities, it shall take measures such as warning, restricting functions, suspending or terminating the provision of services, and shall keep records and report to the relevant competent department in a timely manner.
Article 13 — Where a provider provides generative AI services, it shall sign a service agreement with the user to clarify the rights and obligations of both parties. The provider shall keep training records, generate records, and user information records for at least six months.
Article 14 — The relevant State departments shall, in accordance with their respective responsibilities, strengthen the supervision and administration of generative AI services. Where generative AI services provided in violation of the provisions of these Measures, the relevant competent departments shall, in accordance with the Cybersecurity Law, the Data Security Law, the Personal Information Protection Law, the Law on Science and Technology Progress, and other laws and administrative regulations, impose penalties. Where there are no provisions in the laws and administrative regulations, the relevant competent departments shall, in accordance with their responsibilities, issue warnings, circulate notices of criticism, and order correction within a time limit. Where correction is refused or the circumstances are serious, the provider shall be ordered to suspend or terminate the provision of generative AI services.
Article 15 — These Measures shall take effect as of August 15, 2023.
Disclaimer: This English translation is provided for informational and reference purposes only. While every effort has been made to ensure accuracy, this is an unofficial translation and may contain errors or omissions. The original Chinese text jointly promulgated by the Cyberspace Administration of China and other relevant departments shall prevail as the authoritative version. Readers should consult qualified legal professionals for advice on specific legal matters. Neither the translator nor the publisher assumes any liability for reliance on this translation.
Free PDF download of the complete article.
