From Privacy Concerns to Governance Gaps: University Students’ Perceptions of AI Risks and Ethics

Abstract

This study aims to raise university students’ awareness of AI-related risks and to develop their capacity to identify the legal, ethical, and societal implications of AI use in real-world settings by using a qualitative and learning-embedded design. Fiftyfour undergraduate students in northern Taiwan participated in a two-hour instructional session on AI information security and ethics, followed by guided online searching, reflective writing, and peer sharing on Padlet. Findings indicate that students primarily interpret AI risks through experience-near harms, with privacy erosion, personal data breaches, and cybercrime. Legal issues were mentioned less frequently and were largely limited to broad references to privacy law, portrait rights, and copyright. Participants described AI’s consequences as interconnected socio-technical effects that reshape cognition, equity, labor structures, and educational interaction. Although AI was recognized for efficiency and instructional support, students emphasized that risks intensify when deployment outpaces ethical reflection, governance, and human oversight. To sum up, this study highlights a gap between students’ intuitive ethical concerns and their understanding of legal and governance frameworks, underscoring the need for context-sensitive, practice-oriented AI education that strengthens governance literacy and responsible judgment.