Concern to Acceptance: A Recurrent Study of Students’ Perspectives on GenAI in Higher Education
This recurrent study of Australian university students reveals a significant shift between 2023 and 2025 from initial concern to pragmatic acceptance of generative AI, characterized by increased usage and comfort alongside persistent ethical and skill-related concerns, ultimately suggesting a new theoretical framework for understanding post-adoption AI behavior in higher education.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the landscape of modern learning, a new force has arrived with the speed of a sudden storm: generative artificial intelligence. These are computer systems capable of creating text, images, and ideas that feel remarkably human. For universities, this technology presents a complex puzzle. On one side, it offers powerful tools to help students write better, research faster, and understand difficult concepts. On the other, it raises deep questions about honesty, the value of a degree, and whether students are actually learning or just copying answers. To make sense of how people react to such a shift, researchers often look at a framework called the Theory of Planned Behavior. This idea suggests that whether a person uses a new technology depends on three things: how they feel about it, what they think their peers and society expect of them, and how confident they feel in their ability to use it correctly. Understanding these feelings is crucial because it helps explain why some students embrace these tools while others resist them, and how those feelings might change as the technology becomes a normal part of daily life.
A team of researchers at Adelaide University set out to watch this change in real time. They wanted to know how the minds of university students shifted between 2023, when generative AI was still a fresh and somewhat mysterious novelty, and 2025, when it had become a routine part of the academic world. They surveyed students twice, first with 353 participants in 2023 and again with 258 participants in 2025. The researchers asked simple but revealing questions: How often do you use these tools? How comfortable do you feel using them? Do you trust them to help you learn, or do you worry they will make you less engaged or dishonest? They also asked students to write down their thoughts in their own words, giving the researchers a chance to hear the nuances of student concerns and hopes that a simple yes-or-no question might miss.
The results show a clear and steady march toward acceptance, but not without hesitation. In 2023, less than a quarter of the students used these tools weekly or daily, and many felt unsure or uncomfortable. By 2025, the picture had changed dramatically. Nearly all the students surveyed had used the technology, with almost half using it every week and a growing number using it every single day. The students who used it felt much more at ease, describing the tools as a normal and helpful part of their study routine. They found the technology useful for everything from fixing grammar and organizing their thoughts to gathering information for research projects. The tools had moved from being a strange experiment to a standard piece of equipment in the student toolkit.
However, this acceptance is not a blind leap of faith. The students who embraced the technology in 2025 were not ignoring the risks; they were weighing them carefully. A significant number of students remained worried about whether the information the AI provided was true, whether their personal data was safe, and whether using these tools was unethical. They feared that relying too much on the machines might stop them from learning how to think for themselves or write well on their own. Yet, despite these fears, the students continued to use the tools. They described a new way of thinking the researchers call "pragmatic acceptance." This is a state where a student acknowledges the dangers and ethical problems of a technology but decides to use it anyway because it is too useful and too common to avoid. They see it as a necessary helper for managing heavy workloads, provided they keep a close watch on the output and do not let it replace their own effort.
The study also revealed that students are becoming more sophisticated in how they use these tools. In 2023, many saw the technology as a quick shortcut to find answers, similar to a faster search engine. By 2025, students described using it as a partner in a longer research process. They talked about using the AI to find starting points for their research, to check if their ideas made sense, or to organize large amounts of information, while still doing the hard work of verifying facts and forming their own arguments. They understood that the machine could generate ideas, but they knew it was their job to judge those ideas. This shift suggests that students are learning to integrate the technology into their workflow rather than letting it take over.
The researchers found that this change was driven by a combination of factors. As students became more familiar with the tools, their confidence grew, making them feel more in control. As they saw their peers and the wider world using the technology, the social pressure to avoid it faded, replaced by a sense that using it was simply part of being a modern student. The study suggests that universities cannot simply ban these tools or pretend they will go away. Instead, the findings point to a need for clear guidance. Students are asking for help on how to use these tools ethically and effectively, wanting to know where the line is between helpful assistance and dishonest conduct. They want to learn how to use the technology to support their critical thinking, not to replace it.
Ultimately, the paper paints a picture of a generation of students who are not rejecting the future, but are learning to navigate it with their eyes wide open. They have moved from a place of uncertainty and occasional use to a place of regular, thoughtful engagement. They accept that generative AI is here to stay and that it offers real benefits, but they remain vigilant about the costs. The story of these two years is not one of students surrendering to machines, but of students finding a practical balance, learning to use a powerful new tool while trying to protect the core skills of learning and thinking that define a university education.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.