Ethical Generative AI Use in Higher Education: An Action Research Study Using the Think Aloud Method

ABSTRACT Background As generative AI becomes increasingly embedded in higher education, it presents significant challenges related to ethical use, academic integrity and students' potential over‐reliance on AI‐generated outputs. The Think Aloud method offers a pedagogical approach for making students' reasoning visible as they interact with generative AI tools. Objective This mixed‐methods action research study explored the role of the Think Aloud method as a pedagogical intervention in supporting students' metacognitive awareness and ethical engagement with generative AI tools in a first‐year entrepreneurship course. Method Data were collected through reflective journals, classroom observations and a post‐intervention survey. The study involved 44 first‐year undergraduate students, with 38 students completing the post‐intervention survey. Qualitative data were analysed thematically, while survey responses were examined using descriptive statistics and an exploratory correlation analysis. Results and Conclusions Findings suggested that Think Aloud supported students' emerging awareness of prompt construction, verification practices and ethical decision‐making when using generative AI. However, students' critical evaluation of AI‐generated content often remained at a surface level, particularly among students requiring additional support. The study indicates that Think Aloud can make students' AI‐related reasoning more visible and can support more intentional engagement with generative AI, but it should not be viewed as a stand‐alone solution. Rather, it appears most effective when embedded within a broader pedagogical framework that includes AI literacy, prompt engineering, verification strategies and evaluative judgement. The study contributes classroom‐based evidence on process‐oriented approaches to ethical generative AI use in higher education and highlights the need for sustained instructional scaffolding to support deeper critical engagement with AI tools.