Unpacking the negative effects of generative AI on student motivation and procrastination

Wiwi Widarsih (Departement of Analytical Chemistry, AKA Bogor Polytechnic I Bogor - Indonesia)
Luthfiana Syachfitri (Department of Psychology, Indonesia University of Education I Bandung - Indonesia Faculty of Psychology, Gadjah Mada University I Yogyakarta - Indonesia)
N. Maheshbabu (Department of Studies and Research in Psychology, Sri Dharmasthala Manjunatheshwara College I Karnataka - India)
Muhammad Luthfan Haziman (Departement of Food Nanotechnology, AKA Bogor Polytechnic I Bogor - Indonesia Department of Agro-Industrial Technology, Faculty of Agro-Industrial Technology, Universitas Padjadjaran I Bandung - Indonesia)

Jurnal Psikologi
Vol 25, No 1 (2026): April 2026

Background: The rapid integration of Generative Artificial Intelligence (GenAI) in higher education has reshaped how students engage with academic work. While GenAI improves efficiency and accessibility, concerns arise regarding its effects on cognitive engagement and self regulation.Purpose: This study examined how Ease of Internet Access (EIA) and Frequency of GenAI Use (FGAI) influence Learning Motivation (LM), with Academic Procrastination (AP) as a mediating variable.Method: A total of 205 undergraduate students from Politeknik AKA Bogor completed standardized questionnaires adapted to GenAI-related learning. Data were analyzed using multiple regression, mediation analysis based on Baron and Kenny’s framework, and multi-group confirmatory factor analysis (MGCFA).Findings: The results showed EIA and FGAI did not significantly predict LM (R² = .006; p > .05). EIA significantly predicted AP (β = .146, p = .039), and AP negatively predicted LM (β = −.603, p < .001). Mediation analysis confirmed a significant indirect effect of EIA on LM through AP (Sobel = −2.075, p = .038). MGCFA supported configural and metric invariance across GenAI-use groups (ΔCFI = .003), with partial scalar invariance achieved.Implication: These findings indicate that digital accessibility may indirectly reduce motivation by increasing procrastination, emphasizing the importance of self-regulation and guided AI integration in higher education.