ETHICAL USE OF ChatGPT IN RESEARCH WRITING

Abstract

The exponential integration of generative artificial intelligence (AI) tools such as ChatGPT into academic contexts has raised critical ethical and pedagogical questions regarding their responsible use in research writing. This study explored the lived experiences of 150 students across secondary, undergraduate, and graduate levels in a Science and Technology University in the Philippines concerning the ethical use of ChatGPT in academic paper writing. Guided by hermeneutic phenomenology, the research sought to interpret how students understand instructor guidance, negotiate trust in AI outputs, and manage tensions between assistance and academic integrity. Data were collected through open-ended survey responses and analyzed thematically using an interpretive approach. Findings revealed five essential structures of experience: (a) holding ethical boundaries, (b) conditional trust through verification, (c) prompting as an acquired skill, (d) tension between helpfulness and dependence, and (e) access-driven workarounds. Participants consistently framed ChatGPT as a supportive yet potentially risky tool, requiring verification, transparent acknowledgment, and sustained authorship of thinking. The study concludes that ethical AI use is not merely rule compliance but an evolving moral and epistemic practice shaped by instruction, infrastructure, and academic culture. This research contributes to the body of knowledge by advancing a phenomenological understanding of ethical AI engagement across academic levels and by conceptualizing responsible AI use as a negotiated practice of boundary-setting, verification, and authorship preservation.

Ähnliche Arbeiten