生成式人工智能的回应性监管架构及其实现路径  

Responsive Regulatory Architecture for Generative Artificial Intelligence and Its Realisation Path

作　　者：任梅 代栋栋 Ren Mei;Dai Dongdong(College of Politics and Public Administration,Tianjin Normal University,Tianjin 300387)

机构地区：[1]天津师范大学政治与行政学院,天津300387

出　　处：《中国行政管理》2025年第2期39-50,共12页Chinese Public Administration

基　　金：国家社会科学基金一般项目“新时代农民合作社政治功能的意蕴、实效及理性回归研究”(编号:22BZZ005);国家自然科学基金项目“‘一网统管’下的社区敏捷-韧性治理研究:理论、机制与实现路径”(编号:72174143);教育部哲学社会科学研究后期资助重大项目“中西比较视野下中国合作社制度研究:理论建构与制度优势”(编号:24JHQ013)。

摘　　要：以ChatGPT为代表的生成式人工智能的诞生,在给人类社会带来众多便利的同时,也引发了诸多技术风险。目前我国虽已初步形成系统化的人工智能监管模式,但仍在监管权分配、监管制度设置以及监管策略运用等方面存在不足。面对开放化、协同化的生成式人工智能发展环境,现行政府监管模式需要作出革新。作为一种新的监管模式,回应性监管强调监管主体的多元性、监管制度的敏捷性以及监管策略的多样性,与生成式人工智能监管有着理论、实践和现实适恰性。因此,本文从主体、制度和策略三个维度构建生成式人工智能回应性监管框架,提出推动多主体监管权责优化配置、构建刚柔并济的监管制度和分级分类的监管策略等实现路径,以提升政府对生成式人工智能风险的应对能力,推动人工智能行业的可持续发展。The birth of generative artificial intelligence represented by Chat GPT,while bringing numerous conveniences to human society,also triggers many technical risks.although China has initially formed a systematic AI regulatory model,there are still deficiencies in the allocation of regulatory power,the setting of the regulatory system and the use of regulatory strategies.In the face of the open and collaborative generative AI development environment,the current government regulatory model needs to be innovated.As a new regulatory model,responsive regulation emphasises the plurality of regulatory subjects,the agility of regulatory systems and the diversity of regulatory strategies,which is theoretically,practically and realistically appropriate for generative AI regulation.Therefore,a responsive regulatory framework for generative AI is constructed from the three dimensions of subject,system and strategy,and a path of realisation is proposed to promote the optimal allocation of regulatory powers and responsibilities among multiple subjects,build a rigid and flexible regulatory system and a hierarchical and classified regulatory strategy,so as to enhance the government's ability to respond to the risks of generative AI and promote the sustainable development of the AI industry.

关 键 词：生成式人工智能 回应性监管 监管架构 政府监管 ChatGPT 

分 类 号：D035[政治法律—政治学]