Artificial intelligence systems become increasingly more prevalent in several aspects of daily life. One area in which integration is especially important is within educational systems. AI offers new ways for students to learn and memorize information, condense and simplify complex topics, and interact with other classmates and teachers. As a team, we were interested in qualifying the impact of this integration on educational outcomes. We hypothesize that students will be enthusiastic about the capabilities of AI systems, but concerned about the accessibility issues that may arise as its use becomes more widespread. To determine answers to such questions, we decided it would be best to analyze the opinions of students already within the university systems experiencing technological adoption firsthand. Therefore, for this project, we chose a survey dataset titled “AI Adoption Usage, Expectation, Attitudes, Perceptions, and Motivations for Learning in Higher Education” which asks several Indonesian university students questions on their perception of the performance of AI systems, challenges they anticipate with its use,  and their motivations for using such systems. This dataset was put together by Universitas Negeri Padang, an Indonesian university, and comprises data from 535 associate or undergraduate students who are enrolled in either public or private universities in 20 provinces in Indonesia. We will argue how the survey responses in this dataset offer insight into Indonesian university students’ perceptions of AI usage and how such systems may exacerbate educational inequalities already prevalent in Indonesian provinces. This public perception may demonstrate the need for a balance between complete technological integration and traditional educational methods, implying that certain risks and harms of AI may outweigh the perceived benefits. 

Literature Review

In recent publications examining the surge of AI learning tools in education, particularly Large Language Models (LLMs), many authors share a profound confidence in AI’s potential to transform educational outcomes, and a similar caution against the danger of over-dependence. For instance, in a study of STEM students at the National University of Science and Technology POLITEHNICA Bucharest, an overwhelming 95.6% of the respondents reported using AI for educational purposes (Vieriu and Petrea). As these tools become more ubiquitous, scholars have shifted their focus towards the intersection of benefits, drawbacks, and practical applications. Research suggests that generational identity, regional social factors, and institution policies significantly dictate how often, effectively, and responsibly university students engage with these technologies (Babu, et al; Mytra, et al.; Sari et al.). These demographic discrepancies raise critical questions regarding the fairness of AI literacy and the “digital divide” in access. While many authors and student surveys maintain an optimistic outlook on AI’s ability to enhance learning (Fauziddin et al.), our team intends to narrow down our scope to the provincial differences of AI usage and apply existing scholarly frameworks to our specific dataset. Given how “little is known” about “teachers’ and students’ attitudes towards generative AI as such technologies are only just becoming widely available and will continue to advance at rapid speeds” (Chan and Tsi) we wish to analyze how the regional differences in Indonesia’s unique organization into 38 provinces influence AI usage and student perspectives on AI.

Significance

Rapid development and adoption of artificial intelligence in recent years has made this topic increasingly important as more students incorporate AI tools into their academic lives. This project examines how Indonesian university students understand AI not simply as a convenient learning aid, but as a force that may reshape job preparation, accessibility, and the traditional classroom environment. We are working on this topic because we want to find out whether students see AI as equally beneficial for everyone, or whether they believe it may deepen existing inequalities between regions, institutions, and levels of technological access. This matters because discussions of AI in education often focus on innovation and efficiency, while paying less attention to who is left behind when access, digital literacy, and institutional support are uneven. By studying students’ own perceptions, our project helps others understand that AI can function as a magnifier: it may improve learning and expand opportunities for some students, while simultaneously increasing barriers for others. In this way, our research contributes to larger conversations about fairness, educational equity, and the future role of technology in universities. Rather than asking whether AI should be used at all, this project asks a more important question: under what conditions does AI genuinely support students, and when does it risk replacing meaningful human instruction or worsening performance and accessibility gaps?