Measures on the Administration of Generative Artificial Intelligence Services of the People’s Republic of China — Full English Translation (2023)

Issued 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

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 on Scientific and Technological Progress of the People’s Republic of China, and other relevant laws and administrative regulations for the purpose of promoting the healthy development and standardized application of generative artificial intelligence, safeguarding national security and the public interest, and protecting the lawful rights and interests of citizens, legal persons, and other organizations.

Article 2 — These Measures 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. For the purposes of these Measures, “generative artificial intelligence” refers to models and related technologies that can generate content such as text, images, audio, and video. Where the state has other provisions on the provision of generative artificial intelligence services such as news, publishing, film and television production, and literary and artistic creation, such provisions shall prevail. These Measures do not apply to the research, development, and application of generative artificial intelligence technology that is not provided to the public within the territory of China by industry organizations, enterprises, educational and scientific research institutions, public cultural service institutions, and relevant professional institutions.

Article 3 — The state shall adhere to the principles of attaching equal importance to development and security and promoting innovation and governance in accordance with the law, and shall adopt effective measures to encourage the innovative development of generative artificial intelligence and implement inclusive, prudent, and classified and graded supervision of generative artificial intelligence services. The state encourages the innovative application of generative artificial intelligence technology in various industries and fields, supports inter-institutional collaborative innovation in the industry, academia, and research sectors, promotes the orderly and open sharing of professional data, and promotes the improvement of the technological level, data quality, and computing power base of generative artificial intelligence.

Article 4 — The provision and use of generative artificial intelligence services shall comply with the requirements of laws and administrative regulations, uphold the correct political direction, public opinion orientation, and value orientation, and shall not generate content that endangers national security and interests, damages the national image, incites subversion of state power or the overthrow of the socialist system, endangers national unity and social stability, promotes terrorism or extremism, advocates ethnic hatred and ethnic discrimination, or contains violence, obscenity, or false and harmful information.

Article 5 — The research and development and provision of generative artificial intelligence services shall respect intellectual property rights and business ethics, uphold honesty and trustworthiness, shall not use algorithms, data, platforms, or other advantages to engage in unfair competition, and shall not infringe upon the lawful rights and interests of others such as portrait rights, reputation rights, honor rights, privacy rights, and personal information rights. Generative artificial intelligence service providers shall take effective measures to enhance the transparency of generative artificial intelligence services and improve the accuracy and reliability of generated content.

Article 6 — The relevant state authorities shall, in accordance with their respective duties, strengthen the administration of generative artificial intelligence services. The cyberspace administration authority shall be responsible for the coordination and overall management of generative artificial intelligence services. Departments such as development and reform, education, science and technology, industry and information technology, public security, radio and television, and press and publication shall, in accordance with their respective duties, strengthen the administration of generative artificial intelligence services.

Article 7 — Generative artificial intelligence service providers shall carry out pre-training, optimization training, and other training data processing activities in accordance with the law, and shall comply with the following requirements: (1) use data and basic models from lawful sources; (2) not infringe upon the intellectual property rights of others; (3) where personal information is involved, obtain the consent of the individual or otherwise comply with the provisions of laws and administrative regulations; (4) take effective measures to improve the quality of training data and enhance the authenticity, accuracy, objectivity, and diversity of training data; and (5) comply with other requirements of the Cybersecurity Law, the Data Security Law, the Personal Information Protection Law, and other laws and administrative regulations.

Article 8 — When labeling data manually in the research and development process of generative artificial intelligence, the provider shall formulate labeling rules that comply with the requirements of these Measures and carry out spot checks and verifications of the labeling results. The provider shall conduct necessary training for the labeling personnel and strengthen the management of the labeling personnel.

Article 9 — Where generative artificial intelligence services are provided, the provider shall bear the responsibility as the producer of network information content, and shall bear the responsibility as the provider of personal information processing in accordance with the law. The provider shall sign a service agreement with the user of the generative artificial intelligence service to clarify the rights and obligations of both parties. Where a generative artificial intelligence service provider discovers that a user uses generative artificial intelligence services to commit illegal acts, it shall take measures such as warning, restricting functions, suspending services, or terminating services in accordance with the law, preserve relevant records, and report to the relevant competent authorities.

Article 10 — Generative artificial intelligence service providers shall take effective measures to prevent discrimination based on ethnicity, religion, national origin, gender, age, occupation, health status, and other characteristics during the process of algorithm design, training data selection, model generation and optimization, and service provision. The provider shall use effective and reliable methods to improve the quality of generated content and reduce the generation of discriminatory content.

Article 11 — Generative artificial intelligence service providers shall respect the lawful rights and interests of others such as portrait rights, reputation rights, honor rights, privacy rights, and personal information rights, and shall not use generated content to infringe upon the lawful rights and interests of others.

Article 12 — Generative artificial intelligence service providers shall clearly label generated content such as images and videos in accordance with the provisions of the Measures on the Administration of Deep Synthesis of Internet Information Services. Where generative artificial intelligence service providers discover or become aware that generated content is illegal, they shall promptly take measures to stop the generation and transmission of such content, eliminate it, and take corrective measures such as model optimization and training. Where users are found to be using generative artificial intelligence services to commit illegal acts, the provider shall take measures such as warning, restricting functions, or suspending services, and report to the relevant competent authorities.

Article 13 — Where generative artificial intelligence services are provided, the provider shall conduct security assessments and algorithm filings in accordance with the requirements of the Cybersecurity Law, the Provisions on the Administration of Algorithmic Recommendation Services for Internet Information, and the Measures on the Administration of Deep Synthesis of Internet Information Services.

Article 14 — The relevant competent authorities shall, in accordance with their duties, conduct supervision and inspection of generative artificial intelligence services in accordance with the law. Generative artificial intelligence service providers shall cooperate with the supervision and inspection carried out by the relevant competent authorities in accordance with the law, provide necessary technical and data support and assistance, and truthfully provide relevant materials and data.

Article 15 — Where generative artificial intelligence services are provided and may affect public opinion attributes or social mobilization capabilities, a security assessment shall be conducted in accordance with the relevant provisions of the state, and algorithm filing procedures shall be completed in accordance with the provisions of the Measures on the Administration of Algorithmic Recommendation Services for Internet Information.

Article 16 — Generative artificial intelligence service providers shall establish and improve complaint and reporting mechanisms, set up convenient complaint and reporting portals, promptly accept and handle complaints and reports from the public on generative artificial intelligence services, and provide feedback on the results of handling.

Article 17 — Where generative artificial intelligence service providers provide services for minors, they shall take necessary measures to prevent minors from becoming addicted to or excessively using generative artificial intelligence services. They shall establish mechanisms for protecting minors in accordance with the law and shall not provide minors with generative artificial intelligence services that may harm their physical and mental health.

Article 18 — Generative artificial intelligence service providers shall protect the security of user input information and usage records in accordance with the law, shall not collect unnecessary personal information, and shall not illegally retain or provide to others user input information and usage records that can identify the user. Generative artificial intelligence service providers shall process personal information in accordance with the law and shall not illegally collect, use, process, or transmit personal information.

Article 19 — Where generative artificial intelligence service providers provide services to overseas parties, they shall comply with the relevant state provisions on cross-border data transfer and security assessment.

Article 20 — The state encourages generative artificial intelligence service providers to adopt methods such as self-testing, third-party testing, and industry self-discipline to improve the quality and security of generative artificial intelligence services. The state supports industry organizations to play a role in self-discipline management, formulate industry standards and codes of conduct for generative artificial intelligence, and promote the healthy development of the industry.

Article 21 — Where a generative artificial intelligence service provider violates the provisions of these Measures, the relevant competent authorities shall, in accordance with their duties, issue a warning, circulate a notice of criticism, and order the provider to make corrections within a prescribed time limit. Where the provider refuses to make corrections or the circumstances are serious, it shall be ordered to suspend the relevant services and shall be subject to a fine of not less than 10,000 yuan but not more than 100,000 yuan imposed in accordance with the law. Where laws and administrative regulations provide otherwise for penalties, such provisions shall prevail. Where the provision and use of generative artificial intelligence services violate the laws and administrative regulations on cybersecurity, data security, personal information protection, and other relevant provisions, penalties shall be imposed in accordance with such laws and administrative regulations. Where such violation constitutes a crime, criminal liability shall be pursued in accordance with the law.

Article 22 — For the purposes of these Measures, “generative artificial intelligence service provider” means an organization or individual that provides generative artificial intelligence services by using generative artificial intelligence technology, including the provision of generative artificial intelligence services by providing programmable interfaces and other means.

Article 23 — Where generative artificial intelligence services are provided involving the generation of content that is false and harmful, or infringes upon the lawful rights and interests of others, penalties shall be imposed in accordance with the provisions of the Cybersecurity Law, the Data Security Law, the Personal Information Protection Law, and other relevant laws and administrative regulations.

Article 24 — These Measures shall come into force on August 15, 2023.