DeepSeek类生成式人工智能嵌入意识形态治理的价值、风险与应对  

The Value,Risks and Countermeasures of Embedding DeepSeek-like Generative Artificial Intelligence into Ideological Governance

作　　者：周利生 刘芳华 ZHOU Lisheng;LIU Fanghua(School of Marxism,Jiangxi Normal University,Nanchang 330022)

机构地区：[1]江西师范大学马克思主义学院,江西南昌330022

出　　处：《南昌大学学报(人文社会科学版)》2025年第2期40-49,共10页Journal of Nanchang University(Humanities and Social Sciences)

基　　金：国家社会科学基金重大项目“新时代新征程中国共产党的使命任务研究”(23ZDA010)。

摘　　要：Deep Seek类生成式人工智能嵌入意识形态治理成为技术应用的特殊场景,阐释其价值、风险和应对显得尤为重要。Deep Seek类生成式人工智能带来的话语主体演变、传播模式革命、呈现方式转型、治理模式创新等成为意识形态治理的重要机遇,而数据质量缺陷、算法操纵、技术压制、信息茧房等无疑成为意识形态治理的挑战。综合运用数据自主、算法规制、技术创新、人机协同的治理策略,可以利用机遇、化解挑战,从而提高意识形态治理效能,营造向上向善的数字空间。The embedding of DeepSeek-like generative artificial intelligence into ideological governance has become a special scenario for technological application,and it is particularly important to explain its value,risks and countermeasures.The evolution of discourse subjects,the revolution in communication models,the transformation of presentation methods,and the innovation of governance models brought about by DeepSeek-like generative artificial intelligence have become important opportunities for ideological governance.However,issues such as data quality defect,algorithm manipulation,technological suppression,and information cocoons undoubtedly pose challenges to ideological governance.By comprehensively applying governance strategies of data autonomy,algorithm regulation,technological innovation,and human-machine collaboration,we can seize opportunities and resolve challenges,thereby improving the effectiveness of ideological governance and creating a digital space that is positive and promotes goodness.

关 键 词：生成式人工智能 DeepSeek 意识形态 治理 

分 类 号：D64[政治法律—政治学]