A corpus-based comparison between American and Chinese higher education institutions’ perceptions towards using Gen-AI

• In Event: Poster Session 1
  In Poster Session: A corpus-based comparison between American and Chinese higher education institutions’ perceptions towards using Gen-AI

Proposal

This poster presents a comparative analysis of how American and Chinese higher education institutions (HEIs) perceive and respond to the rapid development of Generative Artificial Intelligence (Gen-AI). The relevance of this topic to CIES 2025 stems from the increasing global importance of AI in education, policy, and practice. As educational systems worldwide adapt to technological innovations, understanding regional differences in attitudes towards AI is crucial for informing international collaboration and policy development. This study addresses key concerns that resonate with the CIES theme of education and innovation in a digitalized world, offering insight into how HEIs from two major global powers are navigating these changes.

The research is guided by a conceptual framework that draws upon techno-pessimism and techno-optimism. These concepts, reflecting skepticism and enthusiasm towards technological adoption, were central to framing the study’s research questions. Specifically, the framework facilitated an exploration of attitudes regarding academic integrity, ethical concerns, and the potential for innovation. The framework also shaped how the data were interpreted, highlighting the cultural and structural differences between American and Chinese institutions.

Information was sourced from institutional reports, academic publications, and public communications from both countries’ HEIs. In the Chinese context, data were extracted from WeChat-Public-Accounts and official university websites, while the American data primarily came from research publications and media reports. The choice of these sources was informed by the need to capture institutional and public sentiments in a corpus-based analysis. Tools such as Python, AntConc, and sentiment analysis algorithms were employed to analyze keyword frequency, thematic trends, and discourse patterns.

The study used a corpus-based methodology, which supports the generalizability of the findings and provides a clear comparison between the two regions. The keyword analysis, combined with sentiment extraction and thematic discourse analysis, enabled the identification of key concerns and opportunities associated with Gen-AI in both HEIs. The results indicate that American institutions generally embrace Gen-AI with optimism, focusing on innovation and personalized learning, whereas Chinese institutions show greater caution, emphasizing academic integrity and the ethical challenges posed by AI technologies. These differences support the study’s conclusion that Gen-AI adoption is shaped not only by technological capabilities but also by cultural and regulatory environments.

The contribution of this research is both original and timely. It offers a nuanced understanding of the contrasting approaches to Gen-AI in higher education, revealing insights that have not been widely explored in the current literature. The findings contribute to the growing discourse on AI in education, particularly in how it intersects with issues of governance, ethics, and pedagogical innovation. This research highlights the importance of addressing the unique needs of diverse educational systems as they grapple with the challenges and opportunities of AI adoption. This is especially significant for global education policymakers and practitioners as they consider how best to harness AI technologies while maintaining ethical and academic standards.

Author

• Zhangxu Kang, University of Michigan Ann Arbor