Misconceptions, Pragmatism, and Value Tensions: Evaluating Students' Understanding and Perception of Generative AI for Education

TL;DR

This study explores undergraduate students' perceptions and uses of generative AI in education, revealing misconceptions, varied applications, ethical concerns, and potential benefits, with implications for teaching and learning strategies.

Contribution

It provides empirical insights into students' diverse understandings, ethical issues, and aspirations regarding GenAI, highlighting the need for educational guidance and policy considerations.

Findings

Students' definitions of GenAI vary widely and include misconceptions.

Common uses of GenAI are writing and coding.

Students see benefits in summarization and personalized learning, but worry about plagiarism and dependency.

Abstract

In this research paper we examine undergraduate students' use of and perceptions of generative AI (GenAI). Students are early adopters of the technology, utilizing it in atypical ways and forming a range of perceptions and aspirations about it. To understand where and how students are using these tools and how they view them, we present findings from an open-ended survey response study with undergraduate students pursuing information technology degrees. Students were asked to describe 1) their understanding of GenAI; 2) their use of GenAI; 3) their opinions on the benefits, downsides, and ethical issues pertaining to its use in education; and 4) how they envision GenAI could ideally help them with their education. Findings show that students' definitions of GenAI differed substantially and included many misconceptions - some highlight it as a technique, an application, or a tool, while others described it as a type of AI. There was a wide variation in the use of GenAI by students, with two common uses being writing and coding. They identified the ability of GenAI to summarize information and its potential to personalize learning as an advantage. Students identified two primary ethical concerns with using GenAI: plagiarism and dependency, which means that students do not learn independently. They also cautioned that responses from GenAI applications are often untrustworthy and need verification. Overall, they appreciated that they could do things quickly with GenAI but were cautious as using the technology was not necessarily in their best long-term as it interfered with the learning process. In terms of aspirations for GenAI, students expressed both practical advantages and idealistic and improbable visions. They said it could serve as a tutor or coach and allow them to understand the material better. We discuss the implications of the findings for student learning and instruction.