Students’ voices on generative AI: perceptions, benefits, and challenges in higher education

Analysis

Why This Paper Matters

As generative AI tools like ChatGPT rapidly permeate higher education, understanding how students perceive and interact with these technologies is critical for effective integration. This paper by Chan and Hu addresses a timely gap by capturing student voices—a perspective often overshadowed by institutional or faculty-focused discussions. With over 2,400 citations, it has become a foundational reference for educators and policymakers navigating the GenAI landscape.

The study's significance lies in its empirical grounding: a survey of 399 students across disciplines in Hong Kong provides concrete data on familiarity, willingness to engage, and perceived benefits and challenges. By applying John Biggs' 3P model (presage, process, product), the authors link student perceptions to learning approaches and outcomes, offering a theoretical lens that elevates the discussion beyond anecdotal evidence. This framework helps explain why addressing student concerns is essential for fostering effective learning environments.

Moreover, the paper's balanced portrayal of both enthusiasm and apprehension mirrors the broader societal debate around GenAI. It highlights that students are not passive recipients but active stakeholders whose views can shape responsible implementation. As institutions worldwide grapple with policy development, this study offers actionable insights for tailoring GenAI tools to meet student needs while mitigating risks.

Technical Contributions

• Empirical Survey Instrument: Designed and administered a survey capturing students' familiarity, attitudes, perceived benefits, and concerns regarding GenAI in higher education.
• Application of Biggs' 3P Model: Extends the model to the context of GenAI, demonstrating how student perceptions (presage) influence learning approaches (process) and outcomes (product).
• Identification of Key Benefits and Challenges: Systematically categorizes benefits (personalized learning, writing assistance, research support) and challenges (accuracy, privacy, ethics, impact on development and career).
• Policy Recommendations: Provides evidence-based suggestions for policymakers and educators to develop guidelines for responsible GenAI integration.

Results

The survey revealed a generally positive attitude toward GenAI, with students acknowledging its potential for personalized learning support, writing and brainstorming assistance, and research and analysis. However, concerns about accuracy, privacy, ethical issues, and the impact on personal development, career prospects, and societal values were also prominent. While the paper does not report specific quantitative metrics (e.g., percentages), the qualitative findings underscore a nuanced student perspective. The application of the 3P model suggests that these perceptions significantly influence how students approach learning with GenAI, thereby affecting outcomes. The study's large citation count reflects its influence in shaping subsequent research and policy discussions.

Significance

This paper has significantly impacted the AI in education field by centering student voices in the GenAI discourse. It has informed institutional policies and guidelines for responsible AI integration, emphasizing the need to address ethical and privacy concerns alongside pedagogical benefits. Its findings have been widely cited in studies on AI literacy, curriculum design, and academic integrity. By highlighting the importance of student perceptions, the paper encourages a more inclusive and evidence-based approach to adopting GenAI in higher education, ultimately aiming to enhance teaching and learning experiences while safeguarding student development and societal values.