中图分类号:D922;TP399文献标志码:ADOI:10.19358/j.issn.2097-1788.2026.01.009 中文引用格式:盛小宝,刘晋名,姚迁. 基于行政执法责任界分的生成式人工智能内容安全监管研究[J].网络安全与数据治理,2026,45(1):56-63. 英文引用格式:Sheng Xiaobao, Liu Jinming, Yao Qian. Research on the content safety regulation of generative artificial intelligence based on the division of administrative enforcement responsibilities[J].Cyber Security and Data Governance,2026,45(1):56-63.
Research on the content safety regulation of generative artificial intelligence based on the division of administrative enforcement responsibilities
Sheng Xiaobao, Liu Jinming, Yao Qian
Third Research Institute of the Ministry of Public Security
Abstract: The uniqueness of the industrial structure and content generation mechanism of generative artificial intelligence (GenAI) has led to numerous challenges in content security supervision, particularly in the division of administrative enforcement responsibilities. Currently, there is ambiguity and overlap in the roles of public security cybersecurity departments, the Cyberspace Administration, and the Ministry of Industry and Information Technology (MIIT) in GenAI content regulation, leading to unclear enforcement responsibilities, and consequently, ineffective regulation. This issue manifests in mismatches between technical tools and regulatory needs, inadequate applicability of existing laws and regulations, insufficient public awareness, and lack of corporate responsibility. To address these challenges, the regulatory framework must be optimized: First, the clear delineation of responsibilities across enforcement agencies should be prioritized, with the establishment of a robust informationsharing and collaboration mechanism; second, leveraging technology to build a comprehensive, tiered governance framework for content safety; third, encouraging corporate responsibility and collaborative efforts to improve regulatory efficiency. Moving forward, it is crucial to integrate law, technology, and management mechanisms to create a more scientific and standardized regulatory system for GenAI content safety, providing effective institutional safeguards for cybersecurity and social order.