Against the backdrop of the global wave of Artificial Intelligence (AI) technology sweeping through the education sector, traditional high school chemistry experiment teaching faces numerous challenges such as safety concerns, cost, spatiotemporal limitations, and the difficulty in observing and understanding certain experimental phenomena. This study aims to systematically explore how artificial intelligence can empower high school chemistry experiment teaching to break through the bottlenecks of traditional teaching models and enhance teaching quality and learning experience. The research first reviews the core technologies of AI applications in education and relevant educational theoretical foundations, pointing out the deep alignment between AI technology and constructivist learning theory. On this basis, the paper systematically proposes four innovative practice pathways for AI-empowered high school chemistry experiment teaching: constructing intelligent virtual/augmented reality laboratories, achieving intelligent guidance and safety monitoring of experimental processes, enabling intelligent analysis of experimental data and report generation, and supporting personalized experiment design and inquiry-based learning. To integrate theory with practice, this study designs and proposes a P-I-A-E (Preparation & Prediction, Interaction & Inquiry, Analysis & Assessment, Evaluation & Extension) four-stage closed-loop teaching model, and builds a multi-dimensional evaluation index system covering knowledge acquisition, experimental skills, scientific inquiry, and data literacy around this model. Finally, the paper deeply analyzes the potential technical, teacher-related, cost, and ethical challenges in the promotion and application of this model, and provides an outlook on future development trends. The study believes that the deep integration of artificial intelligence will reshape the teaching form of high school chemistry experiments, propelling it towards a direction characterized by intelligence, personalization, precision, and safety.
Ma et al. (Mon,) studied this question.