[Paper Review] Generative AI in Higher Education: A Global Perspective of Institutional Adoption Policies and Guidelines

Yueqiao Jin, Lixiang Yan+3 more|arXiv (Cornell University)|May 20, 2024

Artificial Intelligence in Healthcare and Education21 citations

TL;DR

A global, theory-grounded analysis of 40 universities’ policies on Generative AI adoption in higher education, using Diffusion of Innovations Theory to examine innovation traits, communication channels, and stakeholder roles.

ABSTRACT

Integrating generative AI (GAI) into higher education is crucial for preparing a future generation of GAI-literate students. Yet a thorough understanding of the global institutional adoption policy remains absent, with most of the prior studies focused on the Global North and the promises and challenges of GAI, lacking a theoretical lens. This study utilizes the Diffusion of Innovations Theory to examine GAI adoption strategies in higher education across 40 universities from six global regions. It explores the characteristics of GAI innovation, including compatibility, trialability, and observability, and analyses the communication channels and roles and responsibilities outlined in university policies and guidelines. The findings reveal a proactive approach by universities towards GAI integration, emphasizing academic integrity, teaching and learning enhancement, and equity. Despite a cautious yet optimistic stance, a comprehensive policy framework is needed to evaluate the impacts of GAI integration and establish effective communication strategies that foster broader stakeholder engagement. The study highlights the importance of clear roles and responsibilities among faculty, students, and administrators for successful GAI integration, supporting a collaborative model for navigating the complexities of GAI in education. This study contributes insights for policymakers in crafting detailed strategies for its integration.

Motivation & Objective

• Assess how GAI is aligned with university goals and compatibility with institutional objectives.
• Identify how universities enable trialability and observability of GAI initiatives.
• Examine policy communication channels used to disseminate GAI updates to stakeholders.
• Determine the roles and responsibilities of faculty, students, and administrators in GAI adoption policies.
• Provide policy insights for policymakers crafting comprehensive GAI integration strategies.

Proposed method

• Data were collected from 40 universities across six regions using stratified sampling from QS World University Rankings 2024.
• Official policy documents, guidelines, and statements were collected in English and official languages, translated to English for analysis.
• Thematic analysis was conducted to identify themes related to innovation characteristics, communication channels, and roles/responsibilities.
• Two researchers independently coded documents to develop a codebook, with Cohen’s Kappa used to ensure inter-rater reliability (Kappa > 0.61).
• Analysis focused on three dimensions from Diffusion of Innovations Theory: innovation characteristics (compatibility, trialability, observability), communication channels, and social system (roles/responsibilities).
• Results are organized around three research questions (RQ1–RQ3).

Figure 1: Key themes related the compatibility, trialability, observability of generative AI integration which emerged from the analysed universities’ policies and guidelines.

Experimental results

Research questions

• RQ1 RQ1: How are GAI’s innovation characteristics—compatibility, trialability, and observability—represented in higher education policies?
• RQ2 RQ2: What communication channels are identified in policies for disseminating GAI updates and facilitating discussions among stakeholders?
• RQ3 RQ3: What roles and responsibilities are specified for faculty, students, and administrators in GAI adoption policies?

Key findings

• Universities show a proactive stance toward GAI, emphasizing academic integrity, teaching/learning improvement, and equity.
• Compatibility themes reveal alignment with teaching/learning enhancement (n=38) and concerns about information security/data privacy (n=25).
• Trialability themes stress integrating AI into educational practice with explicit use cases and phased experimentation (n=40).
• Observability themes include ongoing evaluation and public reporting of outcomes (observability n=5, e.g., HKU).
• Communication channels are mainly digital platforms (n=15), with additional interactive sessions and direct channels to stakeholders.
• Roles and responsibilities show clear distributions: Faculty drive curriculum/assessment changes (n=20), Students bear ethical use expectations (n=27), Administrators lead policy development/implementation (n=16).
• Policies emphasize human-centric evaluation, transparency, and ongoing monitoring as AI tools evolve.

Figure 2: Five primary communication channels utilised by higher education institutions to communication information about the adoption policy of generative AI.