中国信通院 | 人工智能数据安全 白皮书

人工智能作为引领新一轮科技革命和产业变革的战略性技术，已成为世界主要国家谋求新一轮国家科技竞争主导权的关键领域。随着政府人工智能战略布局的落地实施，全球人工智能发展正进入技术创新迭代持续加速和融合应用拓展深化的新阶段，深刻改变着国家政治、经济、社会、国防等领域的运行模式，对人类生产生活带来翻天覆地的变化。
数据作为驱动本轮人工智能浪潮全面兴起的三大基础要素之一，数据安全风险已成为影响人工智能安全发展的关键因素。与此同时，人工智能应用也给数据安全带来严峻挑战,如何应对人工智能场景下的数据安全风险日渐成为国际人工智能治理的重要议题。部分国家已率先探索人工智能数据安全风险的前瞻研究和主动预防，并积极推动人工智能在数据安全领域应用，力求实现人工智能与数据安全的良性互动发展。
本白皮书从人工智能数据安全的内涵出发，首次提出人工智能数据安全的体系架构，在系统梳理人工智能数据安全风险和安全应用情况的基础上，总结了国内外人工智能数据安全治理现状，研究提出了我国人工智能数据安全治理建议。

(As a strategic technology leading a new round of scientific and technological revolution and industrial transformation, artificial intelligence has become a key area for major countries in the world to seek dominance in a new round of national scientific and technological competition. With the implementation of the government’s strategic layout of artificial intelligence, the global artificial intelligence development is entering a new stage of continuous acceleration of technological innovation and continuous acceleration of integration and application expansion and deepening, profoundly changing the operation mode of national politics, economy, society, national defense and other fields. Production and life bring about earth-shaking changes.
Data is one of the three basic elements that drive the full rise of this round of artificial intelligence, and data security risks have become a key factor affecting the safe development of artificial intelligence. At the same time, the application of artificial intelligence also brings severe challenges to data security. How to deal with data security risks in artificial intelligence scenarios has increasingly become an important topic of international artificial intelligence governance. Some countries have taken the lead in exploring the prospective research and proactive prevention of artificial intelligence data security risks, and actively promote the application of artificial intelligence in the field of data security, and strive to achieve the benign interactive development of artificial intelligence and data security.
Starting from the connotation of artificial intelligence data security, this white paper proposes the architecture of artificial intelligence data security for the first time. Based on the systematic analysis of artificial intelligence data security risks and security applications, it summarizes the current situation of artificial intelligence data security governance at home and abroad, and researches and proposes Adopted my country's artificial intelligence data security governance recommendations.)

人工智能数据安全白皮书.pdf

2021-8-15 16:49 上传

文件大小: 1.15 MB
下载次数: 0