Highlighted Session: Perils and Promises of Generative AI in Higher Education: Comparative Insights from Mainland China, Hong Kong, Japan, Mongolia, and the U.S.

Group Submission Type: Highlighted Paper Session

Proposal

As Generative Artificial Intelligence (GAI) continues to expand its reach within higher education, it reveals both the shared challenges and unique experiences of students and educators across different global contexts. This session seeks to critically examine the varied impacts of GAI on higher education systems by bringing together four studies that each offer a distinct yet interrelated perspective based on empirical data from Mainland China, Hong Kong, Japan, Mongolia, and the United States. Through a comparative lens, the session highlights the nuanced ways in which social, cultural, linguistic, and regulatory differences shape the adoption, use, and consequences of GAI. By drawing on the Unified Theory of Acceptance and Use of Technology (UTAUT) and frameworks on student agency and digital equity, this session offers multiple perspectives on how GAI influences higher education across diverse contexts and contributes to the broader discourse on education in a digital society.

The first paper, “Navigating Global and Local Generative AI Tools in Higher Education: Perspectives from Chinese Students,” sets the stage by exploring the experiences of Chinese students as they navigate the dual realities of restricted global AI tools and emerging local alternatives. This study applies the UTAUT framework to investigate how performance expectancy, effort expectancy, social influence, and facilitating conditions influence students’ acceptance and use of GAI in a context where access to tools like ChatGPT is limited by local regulations. The findings reveal that cultural norms and regulatory environments significantly shape students’ educational practices, compelling them to creatively integrate available technologies into their learning processes. By focusing on the Chinese context, this paper provides critical insights into how local adaptations of GAI can either mitigate or exacerbate global inequalities in access to educational technologies, offering a broader discussion on the interplay between technology, culture, and national policy in shaping educational outcomes.

Building on this foundation, the second paper, “Exploring Students’ Experiences with Generative AI in Non-English Higher Education Landscapes: Case Studies from Japan and Mongolia,” shifts the focus to the linguistic and cultural challenges that GAI poses in non-English speaking regions. This study emphasizes the comparative aspect by examining how students in Japan and Mongolia engage with GAI tools that are predominantly optimized for English, often leading to reduced effectiveness or even exclusion from the full benefits of these technologies. By utilizing a qualitative approach and the UTAUT framework, the research highlights the significant barriers these students face due to language limitations and cultural nuances, which influence their expectations and experiences with GAI. By comparing the experiences of students in these two countries, the paper further underscores the importance of linguistic inclusivity in the development and deployment of GAI tools in global education. It adds depth to our understanding of how linguistic and cultural contexts can significantly alter the effectiveness of educational technologies, necessitating localized adaptations to meet diverse educational needs.

The third paper, “Students Lead AI: Understanding Student Agency in AI-supported Learning,” focuses on the role of student agency within AI-supported learning environments in Hong Kong. Drawing on theories of student agency and adopting the grounded theory methodology, this study explores how students exercise control and make informed decisions in their interactions with GAI tools. The research identifies four key aspects of student agency: control and redirection of AI outputs, mindful adoption of AI-generated suggestions, external help-seeking to complement AI capabilities, and reflective learning practices. By analyzing these dimensions, the study provides insights into how students in Hong Kong, who operate within a hybrid educational system influenced by both Eastern and Western educational values, navigate the complexities of AI-enhanced learning. While this study uncovers how student agency is shaped by the sociocultural and educational context of Hong Kong, it enriches theoretical understandings of agency by offering a nuanced framework applicable to diverse educational settings globally.

The final paper, “The Impact of Generative AI on STEM Learning and Equity: Digital Literacy Gaps and Ethical Concerns,” brings the discussion to the U.S., where GAI’s role in STEM education highlights critical issues of equity and inclusion. This study integrates the UTAUT framework with concepts of digital equity to examine how GAI is used by students from underrepresented and disadvantaged backgrounds in U.S. STEM fields. The research reveals the dual-edged nature of GAI: while it offers benefits such as personalized learning assistance and overcoming language barriers, it also exacerbates existing inequalities due to persistent digital literacy gaps and uneven access to resources. Echoing findings in Japan and Mongolia, the study also highlights challenges faced by international students in the U.S., who may rely on GAI to navigate language barriers but are particularly vulnerable to its limitations such as cultural insensitivity. Situated within the broader U.S. context of systemic inequality in education, this paper underscores the urgent need for inclusive policies and institutional support to ensure that GAI technologies bridge rather than widen the equity gap in higher education.

Together, these four presentations create a cohesive narrative that emphasizes the value of a comparative approach in understanding the global impact of GAI on higher education. While GAI may appear to be a universal technology, its effects vary significantly when localized. By examining how GAI is adopted and adapted across different social, cultural, linguistic, and regulatory contexts, the session highlights both the common challenges and unique experiences.

This session aligns closely with the CIES 2025 conference theme, “Envisioning Education in a Digital Society”. The comparative lens throughout the session fosters a deeper dialogue on the future of education in an increasingly digital world. This dialogue is crucial for envisioning educational practices and policies that harness GAI’s potential while addressing the diverse challenges it presents across different cultural and regional settings. By integrating theoretical perspectives with empirical insights from diverse contexts, this session makes a significant contribution to the ongoing discourse on the role of digital technologies in shaping equitable and inclusive educational futures globally.

Sub Unit

• East Asia SIG

Organizer

• Ran Liu, University of Wisconsin-Madison

Chair

• Yeji Kim, University of Missouri - Columbia

Individual Presentations

• Navigating Global and Local Generative AI Tools in Higher Education: Perspectives from Chinese Students - Qin Xie, University of Minnesota - Twin Cities; Ming Li, Osaka University; Fei Cheng, Kyoto University

• Exploring Students’ Experiences with Generative AI in Non-English Higher Education Landscape: Case Studies from Japan and Mongolia - Ming Li, Osaka University; Ariunaa Enkhtur, Osaka University; Lilan Chen

• Students Lead AI: Understanding Student Agency in AI-supported Learning - Yun Dai; Sichen Lai

• The Impact of Generative AI on STEM Learning and Equity: Transformative Potential and Challenges for Underrepresented Students - Ran Liu, University of Wisconsin-Madison; Carla Zenobia Glave Barrantes, The University of Wisconsin-Madison; Yue Li, University of Florida