This study aimed to explore online and distance learning (ODL) issues related to higher education during the coronavirus disease (COVID-19) in Nepal. We applied an online survey design with a five-point Likert scale. We surveyed 71 (57 male and 14 female) postgraduate students in science education at a public higher education institution in Kathmandu, Nepal. A Principal Component Analysis identified four major constructs as the components of ODL issues. They are scarcity, efficiency, access, and inconvenience. The results of the Independent Samples Test (t-test and ANOVA) showed that participants' views about scarcity were significantly different across their gender (male and female) and device use (mobile, laptop, and desktop). They were not significantly different in their views about efficiency, access, and inconvenience across gender, device use, hometown, age group, ethnicity, and school type at .05 level of significance.

The integration of AI tools in education is reshaping how students view and interact with their learning experiences. As AI usage continues to grow, it becomes increasingly important to understand how students' perceptions of AI technology impact their academic performance and learning behaviours. To investigate these effects, we conducted a correlational study with a sample of 44 students to examine the relationship between students' perceptions of ChatGPT’s utility—focusing on usage frequency, perceived usefulness, accuracy, reliability, and time efficiency—and key academic outcomes, including content mastery, confidence in knowledge, and grade improvement. Additionally, we explored how these perceptions influence student behaviours, such as reliance on ChatGPT, procrastination tendencies, and the potential risk of plagiarism. The canonical correlation analysis revealed a statistically significant relationship between students' perceptions of ChatGPT's utility and their academic outcomes. Students who viewed ChatGPT as reliable and efficient tended to report higher grades, improved understanding of the material, and greater confidence in their knowledge. Furthermore, the bivariate correlation analysis revealed a significant relationship between dependency on ChatGPT and procrastination (r = 0.546, p < .001), indicating that a higher reliance on AI tools may contribute to increased procrastination. No statistically significant association was identified between ChatGPT dependency and the risk of plagiarism. Future research should prioritize the development of strategies that promote the effective use of AI while minimizing the risk of over-reliance. Such efforts can enhance academic integrity and support independent learning. Educators play a critical role in this process by guiding students to balance the advantages of AI with the cultivation of critical thinking skills and adherence to ethical academic practices.