摘要
当前,职业教育面临着数字化转型的巨大压力,通用大模型作为人工智能的前沿技术,展示了强大的赋能潜力。通用大模型的应用,使职业教育可以实现个性化学习路径的设计和动态教学、自动化生成高质量教学资源与内容共享、职业技能的模拟与评估。然而,通用大模型的应用也伴随着一系列潜在风险。数据隐私和安全问题威胁着学生数据的合法收集、存储和传输;算法偏见导致教育不公平;技术深度应用引发学生技术依赖;技术模式异化传统师生关系。为推动大模型有效应用职业教育,必须加强数据隐私保护,确保数据的合法收集和透明处理;设计算法公平性与透明性,通过多样化的数据集和定期的算法审计来减少偏见;设计以学生为中心的学习活动,维护学生的自主性和独立思考能力;强调技术与情感联结的平衡,以保持师生之间的联系互动。
Currently,vocational education is facing enormous pressure from digital transformation,and the general large model,as a cutting-edge technology of artificial intelligence,demonstrates strong em⁃powerment potential.The application of the general large model enables vocational education to design per⁃sonalized learning paths and dynamic teaching,automate the generation of high-quality teaching resources and content sharing,and simulate and evaluate vocational skills.However,the application of general large model also comes with a series of potential risks.Data privacy and security issues pose a threat to the legiti⁃mate collection,storage,and transmission of student data;Algorithm bias leads to educational inequality;Deep application of technology triggers students'dependence on technology;The technological model alien⁃ates the traditional teacher-student relationship.To promote the effective application of large model in voca⁃tional education,it is necessary to strengthen data privacy protection,ensure the legitimate collection and transparent processing of data;Design algorithm fairness and transparency,reduce bias through diverse da⁃tasets and regular algorithm audits;Design student-centered learning activities to maintain students'auton⁃omy and independent thinking ability;Emphasize the balance between technology and emotional connec⁃tion to maintain the connection and interaction between teachers and students.
作者
何柏略
刘衍峰
He Bailue;Liu Yanfeng
出处
《中国职业技术教育》
北大核心
2024年第28期52-62,共11页
Chinese Vocational and Technical Education
基金
2024年度教育部人文社会科学研究专项任务项目“大学生网络圈群社交行为形成机理及舆情引导机制研究”(项目编号:24JDSZ3004,主持人:曾小娟)。
关键词
通用大模型
职业教育
人工智能
general large model
vocational education
artificial intelligence