大语言模型赋能高校思想政治教育:机遇、风险及应对  

Large Language Models Enabling University Ideological and Political Education:Opportunities,Risks,and Responses

作　　者：周从标 赵钰 周勇 ZHOU Congbiao;ZHAO Yu;ZHOU Yong(School of Marxism,Yangtze University,Jingzhou 434023,China;School of Mathematics and Big Data,Chaohu University,Hefei 238024,China)

机构地区：[1]长江大学马克思主义学院,湖北荆州434023 [2]巢湖学院数学与大数据学院,安徽合肥238024

出　　处：《湖北师范大学学报(哲学社会科学版)》2025年第2期103-111,156,共10页Journal of Hubei Normal University(Philosophy and Social Science)

基　　金：高等教育重大决策部署项目(2022jcbs037);安徽省合肥市巢湖学院新建专业(大数据管理与应用)质量提升项目(2023xjzlts074);安徽省合肥市巢湖学院传统专业(应用统计)改造提升项目(2023zygzts088)。

摘　　要：大语言模型以其强大的自然语言处理能力为高校思想政治教育提供了创新的教学手段和丰富的教学资源,增强了高校思想政治话语传播。同时,还能够辅助生成高质量的研究论文和报告,提高研究者的学术成果产出效率。然而,大语言模型在高校思想政治教育的广泛应用也带来数据偏见、过度依赖及隐私泄露等风险。因此,需要建立完善的数据审核机制,培养学生批判性思维,加强数据保护和隐私安全,为大语言模型赋能高校思想政治教育营造良好的生态环境。With its powerful natural language processing capability,the large language models provide innovative teaching tools and rich teaching resources for ideological and political education,and enhance the dissemination of ideological and political discourse in universities.Meanwhile,they can assist in generating high-quality research papers and reports,and improve the efficiency of researchers’academic output.However,the wide application of large language models in ideological and political education in universities also brings risks of data bias,over-reliance and privacy leakage.Therefore,it is necessary to establish a well-built data review mechanism,cultivate students’critical thinking,and strengthen data protection and privacy security so as to create a favorable ecological environment for large language models to better enable university ideological and political education.

关 键 词：大语言模型 思想政治教育 思想政治话语 意识形态 

分 类 号：G641[文化科学—高等教育学]