基于生成式人工智能的网络认知域安全技术  

Cyberspace Cognitive Domain Security Technology Based on Generative Artificial Intelligence

作　　者：刘超超 陈雪 肖冰 LIU Chaochao;CHEN Xue;XIAO Bing(Chinese Academy of Cyberspace Studies,Beijing 100048,China;Law School,Tianjin University,Tianjin 300054,China;Institutes of Science and Development,Chinese Academy of Sciences,Beijing 100190,China)

机构地区：[1]中国网络空间研究院,北京100048 [2]天津大学法学院,天津300054 [3]中国科学院科技战略咨询研究院,北京100190

出　　处：《信息安全与通信保密》2024年第8期41-48,共8页Information Security and Communications Privacy

基　　金：国家重点研发计划项目子课题(2022YFC330190103)。

摘　　要：近年来,人工智能、大数据分析等前沿技术的应用与发展,尤其是生成式人工智能大模型的快速发展,加快了人—网络平台—物理空间的三元融合发展的趋势,促使物理空间、网络空间和认知空间的交织融合。网络空间认知领域存在恶意信息多、个体信息易泄露、信息传播易操纵等诸多安全风险,极易传导至物理空间引发社会问题。基于“人工智能+”的思想,尝试分析基于生成式人工智能的安全防御技术,为促进未来网络安全产业发展提供新思路。In recent years,the application and development of cutting-edge technologies such as artificial intelligence and big data analysis,especially the rapid breakthrough of generative artificial intelligence large models,accelerate the trend of triple integration and development of human-network platform-physical space,which promotes the intertwining and integration of the physical space,cyberspace and cognitive space.There are many security risks in the cyberspace cognition domain,such as a lot of malicious information,easy leakage of individual information,easy manipulation of information dissemination,etc.,which can be easily transmitted to the physical space and thus cause social problems.Based on the idea of“artificial intelligence+”,this paper attempts to analyze the security defense technology based on generative artificial intelligence,providing new ideas for promoting the development of cyber security industry in the future.

关 键 词：生成式人工智能 网络空间 认知领域 安全防御技术 网络安全产业 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]