A Descriptive Model of AI-Assisted Information Problem-Solving (AI-IPS): Core Competencies, Opportunities, and Challenges in Higher Education

• In Event: AERA Roundtable Session Friday 11:45 am - Gold 4
  In Roundtable Session: AI-Enhanced Problem Solving and Instructional Design Frameworks

Abstract

This study proposes and validates a descriptive model of AI-Assisted Information Problem-Solving (AI-IPS) in higher education. Building on the Internet-based IPS framework, the model reframes inquiry as an iterative, collaborative process between students and generative AI tools. Data from 124 undergraduates—including chatbot logs, essays, and 20 interviews—were thematically analyzed. The study identifies five key steps (defining problems, refining prompts, interpreting outputs, verifying sources, and presenting findings) and three core competencies: human–AI collaboration, independent thinking, and information literacy. AI use enhanced information evaluation and engagement with complex issues but posed challenges such as weak prompting, poor verification, and over-reliance. The AI-IPS model supports curriculum design that promotes ethical, critical, and effective AI integration in academic learning.

Authors

• Xinyan Zhou, The Chinese University of Hong Kong

• SHUHENG SHI, Chinese University of Hong Kong

• Thomas K.F. Chiu, The Chinese University of Hong Kong