国际议程设置博弈:AI视频生成模型冲击国家认知域安全的机制OACHSSCD

AI video generation models are reshaping international competition in the cognitive domain and creating new challenges for national cognitive security.Drawing on an analytical framework that links technological characteristics,agenda-setting,and cognitive security,this study explains how the complexity,coherence,and controllability of AI video generation models influence international agenda-setting through distinct entry points and transmission pathways.These technological characteristics strengthen the agenda-setting capacity of countries with advanced digital infrastructures in attribute and network agendas,while expanding the influence of digitally disadvantaged countries in the object agenda.The resulting redistribution of discursive power intensifies agenda-setting competition among sovereign states.Specifically,deepfake-enabled object agenda competition undermines cognitive resilience in the short term;covert value projection within attribute agendas erodes cognitive stability over the medium term;and network agenda competition reinforces information cocoons that weaken cognitive autonomy in the long term.As these dynamics become more persistent,the coexistence of discursive power equalization and structural asymmetry generates differentiated security risks.While all sovereign states face heightened vulnerability in object agendas,digitally disadvantaged countries are more exposed to volatility and systemic risks in attribute and network agendas.To address these challenges,states should strengthen cognitive resilience through digital literacy,reinforce cognitive stability through digital sovereignty,and safeguard cognitive autonomy through digital diplomacy.