China: Interim Measures for the Management of Generative Artificial Intelligence Services
[Unofficial translation by Digital Policy Alert]
Chapter I General Provisions
Article 1
In order to promote the healthy development and standardised application of generative artificial intelligence (AI), safeguard national security and social public interests, protect the legitimate rights and interests of citizens, legal entities, and other organisations; these Measures are developed 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 Progress of Science and Technology of the People’s Republic of China”, as well as other laws and regulations.
Article 2
These Measures apply to the use of generative AI technology to provide services that generate any text, image, audio, video, or other such content to the public (hereinafter “generative AI services”) within the territory of the People’s Republic of China.
If the government has issued other regulations regarding the use of generative AI services in activities such as news and publication, film and television production, or literary and artistic creation, those regulations shall prevail.
These Measures do not apply to the research, development and application of generative AI by industry associations, enterprises, education and research institutions, public cultural bodies and professional bodies that are not provided to the public within the territory of the People’s Republic of China.
Article 3
The State adheres to the principles of equal emphasis on development and safety, promotion of innovation and governance in accordance with the law, adopts effective measures to encourage the innovation and development of generative artificial intelligence, and implements inclusive and prudent and classification and grading supervision of generative artificial intelligence services.
Article 4
The provision and use of generative AI products and services shall comply with the requirements of laws and regulations, respect social morality and ethics, and observe the following provisions:
(i) Generated contents shall adhere to core socialist values, and shall not include any content that subverts the power of the State, undermines the socialist system, endangers national security and interests, damages national image, promotes the division of the State, undermines national unity and social stability, promotes terrorism, extremism, ethnic hatred, ethnic discrimination, violence, obscenity and pornography, or disseminates false and harmful information
(ii) In the process of algorithm design, training data selection, model generation and optimization, and services provision, effective measures shall be adopted to prevent discrimination based on race, ethnicity, beliefs, nationality, regional origin, gender, age, occupation, or health, among others.
(iii) Intellectual property rights and business ethics shall be respected, trade secrets shall not be disclosed, and monopolistic or unfair competition behaviours caused by taking advantage of algorithms, data, platforms, and others, shall not be practised.
(iv) The legitimate rights and interests of others shall be respected, others’ physical and mental health shall not be endangered, and others’ image rights, reputation, honor, privacy rights, and personal information rights and interests shall not be infringed;
(v) Taking into account the specific service type characteristics, effective measures shall be adopted to improve the transparency of generative AI services and improve the accuracy and reliability of generated content
Chapter II Technology Development and Governance
Article 5
The innovative application of generative AI technology in different industries and sectors shall be encouraged to generate positive, healthy, and uplifting content, explore and optimize application scenarios, and build an application ecosystem.
Collaboration with Industry organizations, enterprises, educational and scientific research institutions, public cultural bodies, and relevant professional bodies shall be supported for generative AI technology innovation, data resource construction, transformation and application, and risk prevention.
Article 6
These Measures encourage independent innovation of basic technologies such as generative AI algorithms, frameworks, chips, and supporting software platforms, conducting international exchanges and cooperation on an equal and mutually beneficial basis, and participating in the formulation of international rules related to generative AI.
The construction of generative AI infrastructure and public training data resource platforms shall be promoted. The collaborative sharing of computing power resources shall be facilitated, and the efficient use of computing power resources shall be enhanced. The orderly opening of public data shall be promoted based on categorization and classification, and high-quality public training data resources shall be expanded. The adoption of secure and trusted chips, software, tools, computing power and data resources shall be encouraged.
Article 7
Generative AI service providers (hereinafter the “Providers”) shall conduct pre-training, optimization training, and other training data processing activities in accordance with the law, and shall comply with the following requirements:
(i) Use data and foundation models from legitimate sources
(ii) If intellectual property rights are involved, do not infrige upon third-party intellectual property rights
(iii) If personal information is involved, obtain the consent of the individual or satisfy other conditions under applicable laws and regulations
(iv) Take effective measures to improve the quality of the training data, and enhance the authenticity, accuracy, objectivity, and diversity of the training data
(v) Comply with relevant provisions of laws and regulations such as 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, and other regulatory requirements of relevant competent authorities
Article 8
If data annotation is used in the research and development of generative AI technologies, the Providers shall formulate clear, specific, and operational annotation rules that comply with the requirements of these Measures, conduct data annotation quality assessment, conduct sampling verification of the accuracy of the annotation content, conduct necessary training for annotation staff, and supervise and train annotation staff to perform standardized annotation operations.
Chapter III Service Specifications
Article 9
Providers bear responsibility as network information content producers and shall comply with network information security obligations in accordance with the law. If personal information is involved, they shall bear the responsibility of a personal information processor and shall comply with the obligation to protect personal information in accordance with the law.
Providers shall sign service agreements with generative AI service users who register their services (hereinafter “Users”), to specify the rights and obligations of both parties.
Article 10
The Providers shall clearly specify and disclose the intended group of Users, circumstances and purposes of their services; guide Users towards a scientific and rational understanding and lawful use of generative AI technology; and adopt effective measures to prevent minors from excessively relying on or becoming addicted to generative AI services.
Article 11
The Providers shall protect User’s input information and usage records in accordance with the law. They shall not collect non-essential personal information, shall not illegally retain input information and use records that can identify Users, and they shall not illegally provide User's input information and usage records to others.
Article 12
The Providers shall mark generated content such as images and videos, in accordance with the Regulations on the Management of Deep Synthesis Internet Information Services.
Article 13
The Providers shall provide secure, stable, and sustained services throughout the course of their services to ensure normal use by Users.
Article 14
If Providers find illegal content, they shall promptly adopt measures such as ceasing generation, ceasing transmission, eliminating content, and performing model optimization training. A report shall be submitted to the relevant authorities.
Article 15
Providers shall establish and maintain a complaint and reporting mechanism, develop a convenient complaint and reporting channel, publish information on the process and feedback time limit, accept and promptly handle complaints and reports and provide feedback on the results.
Chapter IV Supervision, Management and Legal Liability
Article 16
The departments of internet and information, national development and reform, education, science and technology, industry and information technology, public security, radio and television, press and publication, shall strengthen the management of generative AI services in accordance with their legal authority.
Relevant national authorities shall, considering the characteristics of generative AI technology and its service applications in relevant industries and sectors, improve scientific supervision methods that are compatible with innovation and development, and develop categorization and grading supervision rules or guidelines
Article 17
Providers of generative AI services with public opinion attributes or social mobilization capacity shall conduct security assessments in accordance with relevant national regulations, and perform algorithm filing, and modification and cancellation of filing in accordance with the Regulations on the Management of Algorithm Recommendation for Internet Information Services.
Article 18
Users shall have the right to file a complaint and report to the relevant authorities if they find that generative AI services do not comply with the laws, regulations and these Measures.
Article 19
When the relevant authorities conduct supervision or inspection of generative AI services according to their duties, Providers shall cooperate in accordance with the laws, explain the description of the source, scale, type, annotation rules, and algorithm mechanism for training data, as required, and provide the necessary technical, data and other support and assistance.
Relevant institutions and personnel involved in security assessments and supervision and inspection of generative AI services shall keep State secrets, and trade secrets, preserve the confidentiality of personal information and privacy acquired when performing their duties, and shall not disclose or unlawfully provide them to others.
Article 20
If AI services originating from outside the territory of the People's Republic of China but provided to the public in the territory do not comply with laws, regulations, and these Measures, the Cyberspace Administration of China shall notify relevant institutions to employ technical measures and other necessary measures to address the matter.
Article 21
If a Provider violates the provisions of these Measures, the relevant authorities shall impose penalties in accordance with the provisions of 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 Science and Technology Progress of the People’s Republic of China, as well as other laws and regulations. If there are no provisions in the laws and regulations, the relevant authorities shall, in accordance with their duties, issue warnings, publish criticisms, or order rectification within a specified period. Refusal to rectify, or if the circumstances are serious, will result in suspension of the provision of relevant services.
If the violation infringes public security, public security administrative sanctions will be imposed. If the violation constitutes a crime, criminal responsibility shall be investigated according to the law.
Chapter V Supplementary Provisions
Article 22
For the purpose of these Measures, the following definitions apply:
(i) Generative artificial intelligence technology means models and related technology that have the ability to generate content such as text, pictures, audio, and video.
(ii) Generative artificial intelligence service provider means an organization and individual that uses generative artificial intelligence technology to provide generative artificial intelligence services, including by providing artificial intelligence programmable interfaces or other means.
(iii) Generative artificial intelligence service users means an organization and individuals that use generative artificial intelligence services to generate content.
Article 23
If laws and regulations provide that relevant administrative licenses shall be obtained for the provision of generative AI services, the Providers shall obtain such licenses according to the law.
Foreign investment in generative AI services shall comply with the provisions of laws and regulations related to foreign investment.
Article 24
These Measures shall come into force on 15 August 2023.