AI Overview
As generative AI embeds in higher education, risks persist around over-automation, reduced student agency, and hallucinations. A critical gap remains in understanding how students themselves navigate collaboration with AI across diverse educational tasks.
- Question: How do graduate computer science students' preferred levels of AI automation compare to their actual AI use?
- Setting and data: Georgia Tech Online Master of Science in Computer Science (OMSCS) program; 57 surveyed students, with 44 analyzed who actively use AI in coursework.
- Method: Mixed-methods user study using two sequential surveys with Likert scales and open-ended questions.
- Students' desired automation levels exceeded their actual AI usage across most academic tasks, revealing conditional adoption rather than general resistance.
- Saving time and reducing cognitive load dominated students' motivations, but inaccurate information and reduced critical thinking were major task-specific concerns, especially for technical tasks.
- Students requested AI design features prioritizing transparency and verifiability—such as clickable citations, explicit uncertainty communication, and visible reasoning steps—over increased automation.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
As generative AI becomes embedded in higher education, it increasingly shapes how students complete academic tasks. While these systems offer efficiency and support, concerns persist regarding over-automation, diminished student agency, and the potential for unreliable or hallucinated outputs. This study conducts a mixed-methods audit of student-AI collaboration preferences by examining the alignment between current AI capabilities and students' desired levels of automation in academic work. Using two sequential and complementary surveys, we capture students' perceived benefits, risks, and preferred boundaries when using AI. The first survey employs an existing task-based framework to assess preferences for and actual usage of AI across 12 academic tasks, alongside primary concerns and reasons for use. The second survey, informed by the first, explores how AI systems could be designed to address these concerns through open-ended questions. This study aims to identify gaps between existing AI affordances and students' normative expectations of collaboration, informing the development of more effective and trustworthy AI systems for education.
