AI赋能与工程伦理融入的C语言课程思政教学改革研究——基于“提问脚手架 + 思政微案例”的教学创新

AI-Empowered C Language Teaching Reform Integrating Engineering Ethics—Pedagogical Innovation Based on the “Questioning Scaffold + Ideological Micro-Case” Model

摘要: 生成式人工智能(GenAI)的普及，在带来便利的同时也催生了编程学习中的“工具依赖”现象，而工科课程思政长期存在的“植入生硬”问题也未得到根本解决。为此，文章以《C语言程序设计》课程为实践对象，依托超星学习通这一主流教学平台，尝试构建一套名为“工程伦理案例库”与“AI深度交互脚手架”的双路径教学模式。这套模式的核心意图，是将思政教育的焦点从外在的、有时略显牵强的政治隐喻，转向内生于工程实践本身的责任意识、系统安全观念与职业素养。并设计了基于平台的AI辅助代码审查等具体活动，来促成这种转向。为了检验效果，采用了设置对照班的准实验研究方法。一个学期的实践数据显示，实验班的学生在平台内实践任务的完成质量、基于真实情境的伦理判断测试(SJTs)得分，以及AI交互日志所体现出的思维深度等多个维度，均显著优于采用传统教学模式的对照班。本研究或许可以为同行提供一个参考：如何利用现有的数字化教学环境，在人工智能时代更自然、更有效实现工科课程知识传授、能力培养与价值引领的“三位一体”。

Abstract: The proliferation of generative artificial intelligence (GenAI) has fostered a troubling trend of “tool dependency” in programming education, while the persistent issue of the “awkward implantation” of ideological and political education in engineering curricula remains fundamentally unresolved. Addressing these dual challenges, this paper takes the “C Language Programming” course as a practical subject. Leveraging the mainstream Chaoxing Learning Platform, we propose and attempt to construct a dual-path teaching model termed the “Engineering Ethics Case Library” and the “AI Deep Interaction Scaffold”. The core objective of this model is to shift the focus of ideological and political education from external, and at times somewhat forced, political metaphors towards the intrinsic cultivation of responsibility awareness, system security concepts, and professional skills inherent to engineering practice itself. To facilitate this shift, specific platform-based activities such as AI-assisted code review were designed. To rigorously evaluate its effectiveness, a quasi-experimental research method employing a control group was adopted. Data collected over one semester of implementation demonstrates that students in the experimental class significantly outperformed their counterparts in the control class—who followed a traditional teaching mode—across multiple dimensions. These dimensions include the quality of practical task completion within the platform, scores on Situational Judgment Tests (SJTs) based on authentic ethical scenarios, and the depth of thinking reflected in AI interaction logs. This study may offer a referential framework for peers, illustrating how to utilize existing digital teaching environments to achieve a more natural and effective “trinity integration” of knowledge impartation, ability cultivation, and value guidance in engineering courses during the AI era.