生成式人工智能数据跨境流动的安全风险及治理范式  

Security risks and governance paradigm for cross-border data flows of Generative AI

作　　者：姚迁 刘晋名 盛小宝[1] Yao Qian;Liu Jinming;Sheng Xiaobao(Third Research Institute of the Ministry of Public Security,Shanghai 200100,China)

机构地区：[1]公安部第三研究所,上海200100

出　　处：《网络安全与数据治理》2024年第12期80-87,共8页CYBER SECURITY AND DATA GOVERNANCE

基　　金：中央级公益性科研院所基本科研业务费专项资金项目(C24340)。

摘　　要：随着生成式人工智能技术的快速迭代,相关数据跨境流动的频率显著增加。面对生成式人工智能在数据跨境流动中呈现的多样化应用场景及其所伴随的安全风险,应当从技术防控、制度优化与国际合作三个维度来建构生成式人工智能数据跨境流动的安全风险治理范式,在保障生成式人工智能数据跨境流动安全性与规范性的基础上,实现数据经济发展与数据权利保护的平衡,推动全球数字经济的健康发展。As Generative AI(GAI)technology rapidly evolves,the frequency of associated cross-border data flows has significantly increased.In light of the diverse application scenarios and the security risks accompanying these cross-border data movements,it is essential to establish a risk governance framework for GAI data flows that addresses these challenges through three key dimensions:technical safeguards,regulatory optimization,and international cooperation.By ensuring the security and compliance of cross-border data flows involving GAI,this framework aims to balance the advancement of the data economy with the protection of data rights,thereby fostering the healthy development of the global digital economy.

关 键 词：生成式人工智能 数据跨境流动 多样化场景 安全风险 

分 类 号：D922[政治法律—法学] TP399[自动化与计算机技术—计算机应用技术]