Multi-Level Barriers to Generative AI Adoption Across Disciplines and Professional Roles in Higher Education
Authors: Jianhua Yang, Kerem Öge, Adrian von Mühlenen, Abdullah Bilal Akbulut, Tanya Suzanne Carey, Chidi Okorro
Organizations: Warwick Manufacturing Group, The University of Warwick, Coventry CV4 7AL, UK · Department of Politics and International Studies, The University of Warwick, Coventry CV4 7AL, UK · Department of Psychology, The University of Warwick, Coventry CV4 7AL, UK · Birmingham Business School, The University of Birmingham, Birmingham B15 2TT, UK
Generative Artificial Intelligence (GenAI) is rapidly reshaping higher education, yet barriers to its adoption across different disciplines and institutional roles remain underexplored. Existing literature frequently attributes adoption barriers to individual-level factors such as perceived usefulness and ease of use. This study instead investigates whether such barriers are structurally produced. Drawing on a multi-method survey analysis of 272 academic and professional services (PSs) staff at a Russell Group university, we examine how disciplinary contexts and institutional roles shape perceived barriers. By integrating multinomial logistic regression (MLR), structural equation modelling (SEM), and semantic clustering of open-ended responses, we move beyond descriptive accounts to provide a multi-level explanation of GenAI adoption. Our findings reveal clear, systematic differences: non-STEM academics primarily report ethical and cultural barriers related to academic integrity, whereas STEM and PSs staff disproportionately emphasize institutional, governance, and infrastructure constraints. We conclude that GenAI adoption barriers are deeply embedded in organizational ecosystems and epistemic norms, suggesting that universities must move beyond generalized training to develop role-specific governance and support frameworks.
Figures & tables
Question
Options
Note
Independent variables
In which department or unit do you work at the University?
Combo box (dropdown selection plus one line text box)
STEM or non-STEM
Which of the following best describes your current role?
Academic; Professional Services
Job Role
Dependent variables
What are the main barriers to using GenAI in your teaching/work?
Lack of institutional guidance or policies (Ins); Concerns about academic integrity and student misuse (E); Limited personal knowledge or training on AI tools (Ind); Lack of time to explore and implement AI solutions (Ind); Ethical concerns about bias, privacy, or surveillance (E); Uncertainty about AI’s effectiveness in improving learning outcomes (Ind); Technical difficulties or lack of institutional support (Ins); Concerns about AI replacing human elements in teaching (E); Resistance from colleagues or institutional culture (C); Lack of funding or access to appropriate AI tools (Ins); Preference for traditional teaching methods (C)
Select three most important.
Mediator
How would you describe your current level of GenAI literacy?
Likert scale 1–5, no understanding to expert
Literacy
What is your overall view of GenAI in education?
Likert scale 1–5, very negative to very positive
Attitude
I am concerned that GenAI will become a threat to my job in the next 5 years.
Likert scale 1–5, strongly disagree to strongly agree
Job Threat
Table 1: Questions in the staff survey, where dependent variables have options falling into categories of Ind = Individual, Ins = Institutional, E = Ethical, and C = Cultural.
Figure 1: Conceptual model showing interacting mediators linking independent and dependent variables .
Figure 2: Barrier percentages differentiating job roles and disciplines shown in a treemap.
Model
\boldmath− 2 Log Likelihood
LR \upchi 2
df
p
Intercept-only
41.569
Final
33.056
8.512
3
0.037
Table 2: MLR model fit between STEM/non-STEM and barriers, where STEM = 1 is the predictor reference group.
Outcome Category
B
OR = Exp(B)
p
Notes
Ethical
− 0.545
0.58
0.181
Not significant
Individual
− 0.857
0.424
0.035
non-STEM lower odds than STEM
Institutional
− 0.952
0.386
0.024
non-STEM lower odds than STEM
Table 3: STEM group effects by selected barrier category (reference outcome = cultural; reference predictor group = STEM).
Model
\boldmath− 2 Log Likelihood
LR \upchi 2
df
p
Intercept-only
63.516
Final
33.959
29.558
3
<0.001
Table 4: MLR model fit between academic/non-academic and barriers, where academic = 1 is the predictor reference group.
Outcome Category
B
OR = Exp(B)
p
Notes
Ethical
− 0.196
0.822
0.522
Not significant
Individual
0.502
1.653
0.101
Not significant
Institutional
0.788
2.198
0.016
PS staff higher odds than academics
Table 5: Job role group effects by selected barrier category (reference outcome = cultural; reference predictor group = academic).
Figure 3: SEM model structure and path coefficients, where grey indicates non-significant paths and black indicates significant paths.
ID
Size
Description
Representative Phrases
B1
21
Opposed to GenAI integration
‘I will not be integrating GenAI into my teaching’, ‘I do not intend to integrate GenAI’, ‘I am strongly against the integration of GenAI’, ‘No, I don’t plan to integrate AI into my teaching’, ‘I have no desire to embed genAI and support a ban’
B2
20
Copilot-only policy frustration
‘Copilot is the only authorised tool’, ‘Copilot is inferior to ChatGPT/Claude’, ‘No approval for competitor products’, ‘Shifting guidance on data uploads’, ‘Poor quality and hallucinations’
B3
15
Erodes learning and integrity
‘Undermines critical thinking and writing’, ‘Students bypass skills for instant gratification’, ‘Makes cheating hard to detect; devalues degrees’, ‘Produces superficially competent but inaccurate work’, ‘Shortcuts before learning underlying skills’
B4
14
Eroding critical thinking
‘Students becoming too reliant on GenAI’, ‘GenAI obliterating students’ critical thinking skills’, ‘Students skip work trusting AI’, ‘Over-reliance on prompting instead of thinking’, ‘GenAI spreading misinformation as fact’
B5
13
Factual accuracy and reliability
‘AI isn’t always correct!’, ‘When AI gets it wrong.’, ‘factual accuracy in finance’, ‘still checking the work’, ‘misinterpreted core message’
B6
11
Accuracy and trust concerns
‘Some staff are mistrustful of it.’, ‘Not being accurate, so always have to check over what the output is.’, ‘It is not personal, and content may end up all sounding the same.’, ‘I don’t trust it to be either ethical or good enough quality.’, ‘Whether the information delivered is factual.’
Table 6: Top 10 barrier clusters from free-text entries.
Generative artificial intelligence (GenAI) tools are increasingly present in higher education, yet their adoption has been largely student-driven, requiring instructors to respond to technologies already embedded in classroom practices. While some faculty have embraced GenAI for pedagogical purposes such as content generation, assessment support, and curriculum design, others approach these tools with caution, citing concerns about student learning, assessment validity, and academic integrity. Understanding faculty perspectives is therefore essential for informing effective pedagogical strategies and institutional policies. In this paper, we present findings from a focus group study with 29 STEM faculty members at a large public university in the United States. We examine how faculty integrate GenAI into their courses, the benefits and challenges they perceive for student learning, and the institutional support they identify as necessary for effective and responsible adoption. Our findings highlight key patterns in how STEM faculty engage with GenAI, reflecting both active adoption and cautious use. Faculty described a range of pedagogical applications alongside concerns about student learning, assessment, and academic integrity. Overall, the results suggest that effective integration of GenAI in higher education requires rethinking assessment, pedagogy, and institutional governance in addition to technical adoption.
Akila de Silva, Isabel Hyo Jung Song, Hui Yang +1
Department of Computer Science, San Francisco State University, 1600 Holloway Ave, San Francisco, CA 94132 USA
The rapid proliferation of Generative Artificial Intelligence (GenAI) is reshaping pedagogical practices and assessment models in higher education. While institutional and educator perspectives on GenAI integration are increasingly documented, the student perspective remains comparatively underexplored. This study examines how students perceive, use, and evaluate GenAI within their academic practices, focusing on usage patterns, perceived benefits, and expectations for institutional support. Data were collected through a questionnaire administered to 436 postgraduate Computer Science students at the University of Hertfordshire and analysed using descriptive methods. The findings reveal a Confidence-Competence Paradox: although more than 60% of students report high familiarity with tools such as ChatGPT, daily academic use remains limited and confidence in effective application is only moderate. Students primarily employ GenAI for cognitive scaffolding tasks, including concept clarification and brainstorming, rather than fully automated content generation. At the same time, respondents express concerns regarding data privacy, reliability of AI-generated information, and the potential erosion of critical thinking skills. The results also indicate strong student support for integrating AI literacy into curricula and programme Knowledge, Skills, and Behaviours (KSBs). Overall, the study suggests that universities should move beyond a policing approach to GenAI and adopt a pedagogical framework that emphasises AI literacy, ethical guidance, and equitable access to AI tools.
Higher education institutions are increasingly expected to ensure that both students and staff develop Generative AI (GenAI) literacies. In response, they are introducing professional development programs and embedding GenAI skills within student curricula. However, current educational frameworks typically assume a linear progression of GenAI literacy, implying that foundational technical understanding must precede creative application. This paper challenges such an assumption through a psychometric analysis of a taxonomy-based self-assessment instrument (n = 158). We applied Rasch measurement theory and Guttman ordering to map the latent perceived order of difficulty of GenAI skills across students, academics, and professional staff. Results reveal a fundamental divergence in perceived competence profiles: while academics follow a more traditional linear path, students exhibit an "inverted" profile, frequently mastering high-level creation tasks before acquiring foundational conceptual understanding. Furthermore, the correlation of skill difficulty between students and academics was weak (r = 0.188). We argue that this "skill bypass" creates a fragile sense of fluency, where high self-efficacy in prompting masks low literacy in AI mechanics. These findings challenge the "one-size-fits-all" curricula and provide the empirical basis for diagnostic-driven, modular interventions that foster genuine human-AI synergy.
Eduardo Oliveira, Narelle English, Tracii Ryan +4
University of Melbourne, Australia · 2FernUniversität, Germany · University of Eastern Finland, Finland