如何识别与判定人工智能教育应用伦理问题——基于伦理策略形成的视角

How to Identify and Determine Ethical Issues in Educational Application of Artificial Intelligence——Based on Formation of Ethical Strategies

伦理问题的产生在人工智能教育应用中不可避免。如何识别与应对之,是当前研究的热点,对于人工智能与教育融合创新具有显著的伦理导向作用。已有研究多着力于对伦理风险的预警,在一定程度上忽视了对伦理问题的识别与判断,暴露出“泛伦理化”研究趋向,难以形成系统的伦理策略。实际上,只有辨别人工智能教育应用行为的伦理属性,通过伦理判断的分类指引,经由不同伦理分析框架的多样尝试,才能确定适切的伦理意图并形成可行的伦理方案。从人工智能教育应用行为的自知性、自愿性、社会性三个基本原则出发,提出“识别人工智能教育应用伦理问题”的一般方法,并基于目的论、义务论、美德论等规范伦理理论,确定“判断人工智能教育应用伦理问题是否合乎伦理规范”的方案依据,从而阐释人工智能教育应用伦理策略的形成过程,为积极回应“伦理问题识别意识薄弱”“伦理原则照搬难适应”等实践困境提供理论依据与实践路径。

Ethical issues are inevitable in educational application of artificial intelligence. How to identify and deal with them now is a hot topic, which plays a significant ethical guiding role in the integration and innovation of artificial intelligence and education. Most of existing researches focus on the early warning of ethical risks, and neglect the identification and judgment of ethical issues to a certain extent, thus exposing the trend of pan-ethicalization and making it difficult to form a systematic ethical strategy. In fact, only by distinguishing the ethical attributes of the application behavior, classifying guidance of ethical judgment, and by making various attempts in different ethical analysis frameworks, can appropriate ethical intentions be determined and feasible ethical schemes be formed. So this paper firstly puts forward a general method of identifying the ethical issues based on the three basic principles of selfknowledge, voluntariness and sociality, and then determines the basis of judging whether the ethical issues conform to ethical norms based on the normative ethical theories such as teleology, obligation theory and virtue. By explaining the formation process of the ethical strategies, it provides a theoretical basis and a practical path to actively respond to the practical dilemmas such as "weak awareness of identifying ethical issues" and "difficulty in copying ethical principles".