Standards and Regulation: Cyberspace Administration of China
Standards and Regulation · Cyberspace Administration of China. Formal documents on generative AI management and filing. This source has 5 entries admitted under this category's review policy; each links directly to the original.
Authority tiers in this category: A1.
Provisions on the Administration of Algorithmic Recommendations for Internet Information Services
Sets requirements for algorithmic recommendation services on information content, user rights, filing, security assessment and supervision.
Evidence use: Supports current regulatory requirements for algorithmic recommendation services in China.
Limitations: Applies only to algorithmic recommendation internet information services within mainland China. Must be read with higher-level laws and later regulations.
Related concepts: AI Governance and Regulation
Link last checked: 2026-09-24
Provisions on the Administration of Deep Synthesis of Internet Information Services
Sets requirements for deep synthesis services on data, labeling, security assessment, filing and user management.
Evidence use: Supports current obligations and covered entities for deep synthesis services in China.
Limitations: Limited in both geographic scope and service type. Should be applied together with the new rules on labeling AI-generated content.
Related concepts: AI-generated Content Detection and Watermarking · Voice Cloning · AI Governance and Regulation
Link last checked: 2026-09-24
Interim Measures for the Administration of Generative AI Services
Sets obligations for public-facing generative AI services on training data, content safety, personal information, labeling and service governance.
Evidence use: Supports current regulatory requirements for generative AI services in China and the explicit exemptions.
Limitations: Does not apply to R&D uses that do not serve the public in mainland China. Must be read with filing, security assessment and other laws.
Related concepts: AI Governance and Regulation · Training Data Governance · Privacy and Data Compliance
Link last checked: 2026-09-24
Measures for Labeling AI-Generated and Synthetic Content
Sets rules for explicit and implicit labeling of AI-generated and synthetic content, verification during distribution and platform responsibilities.
Evidence use: Supports who must label, the technical methods, effective date and platform obligations.
Limitations: Must be read with the accompanying mandatory national standard GB 45438-2025. Specific technical requirements differ by content type.
Related concepts: AI-generated Content Detection and Watermarking · AI Governance and Regulation
Link last checked: 2026-09-24
Interim Measures for the Administration of Anthropomorphic AI Interaction Services
Regulates AI services with anthropomorphic expression and interaction, covering identity disclosure, addiction risk, protection of minors and content governance.
Evidence use: Supports current Chinese regulatory obligations for anthropomorphic AI interaction services.
Limitations: Applies only to anthropomorphic interactive services as defined in the measures. Applies together with existing rules such as those on generative AI.
Related concepts: AI Governance and Regulation · Human-in-the-loop (HITL)
Link last checked: 2026-09-24