期刊文献+

生成式人工智能敏捷治理的实现路径 被引量:1

Pathways to Implementing Agile Governance on Generative Artificial Intelligence
在线阅读 下载PDF
导出
摘要 生成式人工智能引发的技术内生风险、个人和企业侵权风险、国家和社会安全风险、智能意识失控风险等使得“科林格里奇困境”再度凸显。敏捷治理模式的柔韧性、流动性、灵活性等优势,有助于抵御生成式人工智能在多模态、涌现性和通用性特征下引发的诸多不确定风险,被世界各国广泛推崇。目前,我国虽已形成了以治理理念、治理对象、治理工具、治理主体为主要内容的敏捷治理基础框架,但相关治理方案仍“不够敏捷”。进一步完善敏捷治理,需从理论向实践转变、从分散向系统转变、从笼统向精细转变,以前瞻性、战略性、科学性、合理性的思维,系统性刻画敏捷治理范式革新,形成治理理念动态平衡、治理工具灵活柔韧、治理对象和场景分层分级、治理主体协同互动的治理新生态,构建全链条、全覆盖、全场景、全流程的敏捷治理新体系。 Risks triggered by generative AI,including technology-endogenous risks,personal and corporate infringement risks,national and social security risks,and risks of uncontrolled AI consciousness,have contributed to the prominence of the"Collingridge Dilemma".Owing to its merits of adaptability,fluidity and flexibility,the agile governance model helps counteract the various uncertain risks brought about by generative AI in terms of its characteristics of multimodality,emergence and generality,thus being widely advocated around the world.At present,although China has developed a basic framework for agile governance with governance concepts,objects,tools and subjects as its core components,its relevant governance solutions remain insufficiently agile.To further refine agile governance,it is necessary to shift from theory to practice,from fragmentation to systematization,and from generalization to refinement.Guided by forward-looking,strategic,scientific,and rational thinking,the paradigm innovation of agile governance should be systematically elaborated,with the goal to foster a new governance ecosystem characterized by dynamically balanced governance concepts,flexible governance tools,stratified and graded governance objects and scenarios,and collaborative interactions among governance subjects,all of which will ultimately establish an agile governance system that covers the entire chain,full coverage,complete scenario and total process in a comprehensive manner.
作者 赵宇健 Zhao Yujian(School of Law,Beijing Normal University,Beijing 100875,China)
出处 《长沙理工大学学报(社会科学版)》 2025年第5期138-148,共11页 Journal of Changsha University of Science and Technology(Social Science)
基金 国家社会科学基金重大项目(23&ZD156)。
关键词 生成式人工智能 敏捷治理 分层分级 动态政企 generative artificial intelligence agile governance stratification and classification dynamic government and enterprise
  • 相关文献

参考文献18

二级参考文献277

共引文献695

同被引文献14

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部