基于“技术道德化”理论的生成式人工智能教育应用潜能与风险研究  

Study on the Potential and Risk of Educational Application of Generative Artificial Intelligence Based on Moralizing Technology Theory

作　　者：童慧[1] 杨彦军[2] TONG Hui;YANG Yanjun(Department of Education,Jiangxi Science and Technology Normal University,Nanchang Jiangxi 330038;Institute of Education Development,Nanchang University,Nanchang Jiangxi 330031)

机构地区：[1]江西科技师范大学教育学部,江西南昌330038 [2]南昌大学教育发展研究院,江西南昌330031

出　　处：《电化教育研究》2024年第7期12-18,共7页E-education Research

基　　金：江西省社会科学“十四五”基金项目2024年度课题“基于‘技术道德化’视角的生成式人工智能应用潜能与风险研究”(项目编号:24JY17);江西科技师范大学2022年度教学改革课题“混合式课程中过程性学习评价改革与实践”(课题编号:202/6000011842)。

摘　　要：随着生成式人工智能的快速发展及其在教育领域的广泛应用,国内外研究者围绕它及其背后的大语言模型教育应用问题展开了热烈的讨论,但大部分研究因缺乏对生成式人工智能技术原理的了解而在问题探讨中存在伊莉莎效应。研究首先在对生成式人工智能技术学原理进行深入剖析的基础上,从维贝克“人—技杂合”视角对生成式人工智能的本质展开技术哲学分析;其次,基于“技术道德化”理论分析了生成式人工智能教育应用的四大潜能和面临的五大风险挑战;最后,提出基于“人—技杂合”的思想从“设计者”和“使用者”两种视角构建“双向奔赴”的全球人工智能治理体系,将成为未来探索的重要方向。With the rapid development of Generative Artificial Intelligence(GAI)and its wide application in the field of education,researchers at home and abroad have launched a heated discussion about it and the educational application of large language model behind it.However,the Eliza effect exists in most studies because of the lack of understanding of the principle of GAI technology.This study firstly analyzes the essence of GAI from the perspective of Verbeke's"human-technology hybrid"on the basis of an in-depth analysis of the principle of GAI technology,and then elaborates four major potentials and five major risks and challenges faced by the application of GAI in education.Finally,based on the idea of"human-technology hybrid",a global AI governance system is proposed from the two perspectives of"designer"and"user",which will become an important direction for future exploration.

关 键 词：生成式人工智能 技术道德化 大模型 人—技杂合 应用风险 

分 类 号：G434[文化科学—教育学]