USE OF ARTIFICIAL INTELLIGENCE TOOLS IN THESIS WRITING: A MIXED-METHODS ANALYSIS OF THEIR IMPACT ON RESEARCH QUALITY, ORIGINALITY, AND ACADEMIC INTEGRITY AMONG GRADUATE STUDENTS

Abstract

Background: The expansion of artificial intelligence (AI) in academic writing has essentially transformed how one develops the thesis, providing unparalleled assistance but, at the same time, provoking the conventional ideas of innovation and academic honesty.

Objective: This study critically examined the impact of AI-assisted writing tools on research quality, originality, and academic integrity among graduate students.

Methods: The sequential explanatory design based on mixed methods was adopted. A questionnaire was used to gather quantitative data on 312 graduate students of various disciplines, and then qualitative semi-structured interviews were performed with 20 purposely sampled participants. The multiple regression modeling and rigorous thematic analysis were used to analyze the data, as they guaranteed methodological triangulation.

Results: The use of AI tools was very common (78.5%), with the majority of them used to synthesize literature, paraphrase, and refine the language. Regression analysis showed that the significant positive effect was on the perceived quality of the research (0.42, p < 0.001), especially on the coherence, academic tone, and productivity. Nonetheless, greater reliance on AI showed a high negative correlation with originality (= -0.31, p < 0.01). It is worth noting that 64.1% of the participants said there was ambiguity when it comes to ethical standards, and 37.8% acknowledged unreported AI-aided content generation. Three overriding themes were identified with the help of qualitative findings, including AI as an intellectual enhancement tool, the loss of autonomous scholarly voice, and critical institutional governance gaps.

Conclusion: Although AI tools have a significant beneficial impact on the quality and efficiency of the research, they pose a threat to originality and academic integrity when they are not used in a controlled manner. The paper highlights the fact that it is critically important to have thorough institutional policies, ethical guidelines, and the integration of AI literacy to protect academic integrity in the growing digital research environment.