This study investigates how AI ethics, ChatGPT usage habits, and memory loss influence learning motivation and learning performance in higher education. The research addresses growing concerns about cognitive and ethical implications of AI tool usage among students, especially in relation to motivation decline and learning outcomes. Although previous studies highlight the benefits of AI tools in enhancing learning, few have explored the negative cognitive and ethical consequences of overuse. This paper fills the gap by examining how learning motivation mediates the relationship between AI-related factors and learning performance through the lens of Self-Determination Theory (SDT). A quantitative, cross-sectional design was employed using purposive sampling. A total of 539 university students who have experience using ChatGPT in academic contexts participated by completing an online questionnaire. Data were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). This study contributes to the literature by integrating AI ethics, usage patterns, and cognitive outcomes into a unified model of academic motivation and performance. It extends Self-Determination Theory into AI-assisted learning environments and emphasizes learning motivation as a psychological bridge linking ethical and cognitive concerns to learning success. AI ethics, ChatGPT usage habits, and memory loss each have a significant im-pact on both learning motivation and learning performance. Learning motiva-tion serves as a mediating variable in the relationship between all three predic-tors and academic outcomes. Among the predictors, ChatGPT usage habits emerged as the strongest positive influence, whereas memory loss exhibited the most pronounced negative effect. Ethical concerns demonstrated a modest yet statistically significant positive effect, particularly when internalized as responsi-ble academic conduct. Future research should explore longitudinal effects of AI tool use on motivation and cognition, test other theoretical frameworks such as TAM or Cognitive Load Theory, and examine new variables like AI literacy, digital well-being, and academic resilience.
Copyrights © 2025