预见性情报视角下生成式人工智能安全风险感知实践探究  

Exploring the Generative AI Security Risk Perception Framework from the Perspective of Predictive Intelligence

作　　者：罗梓超 李荣 刘如 吴晨生 Luo Zichao;Li Rong;Liu Ru;Wu Chensheng(Beijing Institute of Science and Technology Information,Beijing 100044;Chinese Academy of Science and Technology for Development,Beijing 100038)

机构地区：[1]北京市科学技术研究院科技情报研究所,北京100044 [2]中国科学技术发展战略研究院,北京100038

出　　处：《情报理论与实践》2025年第4期64-71,共8页Information Studies:Theory & Application

基　　金：北京市科学技术研究院创新类课题“新时代北京高精尖产业创新生态研究”;北京市科学技术研究院创新类课题“基于人工智能技术的北京市科技情报平台优化提升”;北京市科学技术研究院创新类课题“基于情报大模型的科技情报服务平台集成与示范”的成果

摘　　要：[目的/意义]探索生成式人工智能安全风险预见性情报工作框架,为我国情报机构提供借鉴和启示。[方法/过程]通过文献回顾当前生成式人工智能安全风险感知研究现状,归纳生成式人工智能安全风险发展趋势,分析当前生成式人工智能风险治理的被动性。在此基础上,论述了预见性情报工作相关机制、流程,探索性构筑生成式人工智能安全风险感知框架。[结果/结论]预见性情报工作嵌入生成式人工智能安全风险感知是一种前瞻性的策略,其嵌入机理包含多个相互关联的层面,广泛性、动态性、预测性作为关键支撑要素,可以有效应对当前生成式人工智能安全风险趋势及演变特性,弥补传统情报中的非线性突发事件的感知不足,进而改善当前生成式AI安全风险治理的被动局面。[Purpose/significance]Explore the working framework of predictive intelligence on the security risks of generative artificial intelligence,so as to provide references and inspirations for China’s intelligence agencies.[Method/process]Through literature review to understand the current research status of the perception of the security risks of generative artificial intelligence,summarize the development trends of the security risks of generative artificial intelligence,and analyze the passivity of the current risk governance of generative artificial intelligence.On this basis,discuss the relevant mechanisms and processes of predictive intelligence work,and construct an exploratory framework for the perception of the security risks of generative artificial intelligence.[Result/conclusion]The embedding of predictive intelligence work into the perception of the security risks of generative artificial intelligence is a forward-looking strategy.Its embedding mechanism involves multiple interrelated levels.Universality,dynamism,and predictability,as the key supporting elements,can effectively cope with the current trends and evolutionary characteristics of the security risks of generative artificial intelligence,make up for the insufficient perception of non-linear emergencies in traditional intelligence,and thus improve the passive situation of the current governance of the security risks of generative AI.

关 键 词：生成式人工智能 人工智能安全风险 非线性突发事件 预见性情报 风险感知 

分 类 号：TP1[自动化与计算机技术—控制理论与控制工程]