生成式人工智能背景下的跨境数据安全规制——基于DeepSeek、ChatGPT等主流AI的思考  

Analysis of Cross-border Data Security Regulation in the Generative AI Context——Based on the Application of Mainstream AI Models such as DeepSeek and ChatGPT

作　　者：张亮[1] 陈希聪 ZHANG Liang;CHEN Xi-cong

机构地区：[1]中山大学法学院,广东广州510006

出　　处：《湖北大学学报(哲学社会科学版)》2025年第2期120-128,F0003,共10页Journal of Hubei University(Philosophy and Social Science)

基　　金：国家社会科学基金项目“粤港澳大湾区法治合作的示范法路径研究”(24BFX017)。

摘　　要：生成式人工智能作为新质生产力,其迭代发展须依赖数据要素的自由流动及合理配置。跨境数据流动与生成式人工智能已形成协同共生的客观事实,但数据要素依其分类分级引致程度不一的跨境数据安全问题。在耦合生成式人工智能技术逻辑与特性的背景下,亟需规制措施确保国家、组织及个人数据在生成式人工智能发展中处于被有效保护及合法利用的安全状态。观之主要国家已出台的生成式人工智能规制方案,由于国家间的利益关切需求不同,生成式人工智能技术发展进程中的跨境数据安全规制尚存在无法弥合的差异。以生成式人工智能为标志的新一代产业革命叠加全球人工智能治理规则变革的浪潮下,我国应从总体国家安全观出发,遵循“数据本位—技术本位—合作本位”的规制进路,就生成式人工智能中的跨境数据安全尽快建立符合本土实际且引领国际的规制架构,为构建生成式人工智能全球治理体系贡献中国智慧。As a new quality productivity,the iterative development of generative artificial intelligence(AI)hinges on the seamless flow and rational allocation of data elements.The synergy and symbiosis between cross-border data flows and generative AI have become an established reality.However,the classification of data elements introduces varying degrees of cross-border data security challenges.In light of the coupled technological logic and characteristics of generative AI,regulatory measures are imperative to ensure that national,organizational,and personal data remain in a state of effective protection and legitimate use during the development of generative AI.Despite the regulatory frameworks introduced by major countries for generative AI,unbridgeable differences in cross-border data security regulations persist due to divergent interests,concerns,and needs among nations.Amidst the wave of the new industrial revolution driven by generative AI and global AI governance regulation,China should adopt a holistic national security perspective,follow a regulatory approach encompassing“data standards,technology standards,and cooperation”,and promptly establish a regulatory system that aligns with local realities while leading international efforts in cross-border data security regulation for generative AI.This will contribute Chinese wisdom to the construction of a global governance system for generative AI.

关 键 词：生成式人工智能 数据安全 数据跨境 新质生产力 DeepSeek ChatGPT 

分 类 号：DF90[政治法律—国际法学] D922.29[政治法律—法学]