Multi-Level Barriers to Generative AI Adoption Across Disciplines and Professional Roles in Higher Education
This study utilizes a multi-method analysis of 272 staff at a Russell Group university to demonstrate that barriers to Generative AI adoption are structurally produced and vary systematically by role and discipline, with non-STEM academics primarily citing ethical concerns while STEM and professional services staff emphasize institutional and infrastructure constraints.
Original paper licensed under CC BY 4.0 (http://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
Imagine a university not as a single building, but as a massive, bustling city with different neighborhoods. In this city, there are two main types of residents: the Scholars (teachers and researchers) who live in the "Academic District," and the Operatives (administrative and support staff) who run the "City Services" district.
Recently, a new, powerful tool called Generative AI (like a super-smart robot assistant) arrived in this city. Everyone is excited, but also confused. Some people are eager to use it, while others are holding back.
This paper is like a detective story that asks: "Why are some people in this city more hesitant to use the robot assistant than others? Is it because they are just scared of new tech, or is it because their specific neighborhood has different rules and problems?"
Here is the simple breakdown of what the researchers found:
1. The Old Theory vs. The New Discovery
The Old Theory: Most people thought the hesitation was personal. They believed, "If we just give everyone a training class, they will learn to love the robot." It was like thinking everyone just needs a driving lesson to feel comfortable behind the wheel.
The New Discovery: The researchers found that the hesitation isn't just about personal fear; it's about where you live in the city. The barriers are "structural," meaning they are built into the different neighborhoods themselves.
2. The Two Different Neighborhoods (Disciplines)
The researchers split the Scholars into two groups: STEM (Science, Tech, Engineering, Math) and Non-STEM (Arts, Humanities, Social Sciences).
- The Non-STEM Neighborhood (The Writers & Thinkers):
- The Problem: They are mostly worried about ethics and cheating.
- The Metaphor: Imagine a group of painters. They are terrified that the robot will steal their brushstrokes, forge their signatures, or make their art look fake. They worry that if students use the robot, they won't learn how to think or write for themselves. Their barrier is moral: "Is this right? Is this fair?"
- The STEM Neighborhood (The Builders & Coders):
- The Problem: They are mostly worried about tools and rules.
- The Metaphor: Imagine a group of engineers. They aren't worried about the robot's soul; they are worried that the robot doesn't have the right blueprints, or that the city won't let them buy the specific software they need. Their barrier is practical: "Do we have the budget? Is the software approved?"
3. The Two Different Jobs (Roles)
The researchers also looked at the difference between Teachers and Administrative Staff (Professional Services).
- The Teachers: They see the robot as a pedagogical risk. They worry about how it changes the classroom, how it affects student learning, and whether it breaks the rules of academic honesty.
- The Administrators: They see the robot as a governance headache. They are the ones holding the keys to the city gates. They worry about data privacy laws, licensing fees, and whether the robot will break the city's security systems.
- The Metaphor: If the robot is a new type of vehicle, the Teachers are worried about how it affects the passengers (students). The Administrators are worried about whether the vehicle has a valid license, insurance, and if it's allowed on the road at all.
4. The "City Map" (The Data)
The researchers surveyed 272 people and used three different "lenses" to look at the data:
- The Survey (The Map): They asked people to pick their top three worries.
- The Math (The Compass): They used complex statistics to prove that your job title and your subject area predict what your worries are.
- The Word Cloud (The Voices): They let a computer read thousands of written comments and grouped them by theme.
What the voices said:
- Some people said, "I will never use this!" (Pure resistance).
- Some said, "The university only allows one specific tool, and it's bad!" (Policy frustration).
- Some said, "Students will stop thinking for themselves!" (Integrity fears).
- Some said, "We don't have the money to buy the good version!" (Infrastructure issues).
5. The Big Takeaway
The main lesson of this paper is that you cannot solve this problem with a "one-size-fits-all" training session.
If you try to teach a group of worried writers how to use a robot by showing them a spreadsheet, they won't care. If you try to teach a group of worried administrators about "creative writing," they will be bored.
The Solution:
Universities need to build different roadmaps for different neighborhoods.
- For the Writers, they need clear rules about honesty and how to keep human creativity alive.
- For the Builders, they need better software access and clear technical guidelines.
- For the Administrators, they need clear legal frameworks and budget approvals.
In short: The robot is here to stay. But to make it work, the university can't just treat everyone the same. They have to understand that the "barriers" people face are actually just the different walls and fences of their own specific neighborhoods.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.