Exploring Instructional Design Strategies for Utilizing Generative AI in Early Childhood Teacher Training Programs

This study aims to explore the instructional design approach utilizing generative AI (such as ChatGPT) in early childhood teacher training programs and examine its feasibility and applicability. The research was conducted through literature review, two rounds of expert focus group interviews (FGI), and the development of an instructional design model. The findings indicate that AI-based instructional design should be systematically structured, incorporating adaptive learning, AI-driven feedback and assessment, and intelligent learning assistants (chatbots). AI can serve as a tool to complement the role of instructors and support personalized learning. However, limitations such as the reliability of AI feedback and disparities in AI utilization skills also exist. To address these issues, it was concluded that maintaining the traditional role of instructors and adopting a hybrid assessment model that combines AI evaluation, peer assessment, and instructor evaluation is necessary. Based on the ADDIE instructional design model, a four-week AI-integrated instructional design framework was proposed, detailing AI-assisted collaborative learning and assessment methods. The study concludes that AI-based instructional design has the potential to improve teaching quality and enhance teachers' instructional planning skills in early childhood teacher training programs.