Using Artificial Intelligence to Boost Creativity in Problem-Based Learning

Abstract

The increasing presence of generative artificial intelligence (AI), particularly Generative Pre-trained Transformer (GPT) tools, has prompted renewed attention to how such technologies may support creativity and problem-solving in secondary education. This action research investigates the impact of guided prompt engineering and structured AI-assisted learning activities on Secondary 2 students’ creativity and problem-solving within a problem-based learning (PBL) framework. The study was conducted in a Hong Kong secondary school across four Secondary 2 classes, where students collaboratively designed AI-supported web applications to address authentic learning challenges. A mixed-methods research design was employed, incorporating pre- and post-surveys, product-based assessments, classroom observations, and group interviews to capture both quantitative trends and qualitative insights. Findings indicate that while students initially demonstrated varied levels of familiarity and confidence with GPT tools, structured instructional guidance in prompt engineering significantly enhanced their creative engagement, idea generation, and ability to translate abstract concepts into practical solutions. Students reported increased motivation, deeper engagement in design and coding tasks, and greater confidence in iterative problem-solving and collaboration. Although concerns regarding over-reliance on AI and technical limitations were identified, the results suggest that GPT can function effectively as a cognitive partner when its use is carefully scaffolded. This study contributes evidence-based insights into responsible AI integration and offers practical implications for educators and school leaders seeking to foster creativity and higher-order thinking in AI-enhanced learning environments.