Seocho High School, Seocho-gu, Seoul, Republic of Korea

Abstract: This paper addresses three critical challenges in high school physics education in Korea: the abstractness of concepts (difficulty in visualizing invisible physical phenomena), the passivity of learning (students as passive recipients of formulas), and the lag in evaluation (over-reliance on summative written tests). Grounded in the Korean "2022 Revised Physics Curriculum" and the "AI Digital Textbook" initiative, this study proposes an innovative instructional model: the "AI-Empowered Three-Stage Modeling Cycle" (Prediction-Modeling-Verification). Through a one-semester action research project involving 120 students in Grade 11 at Seocho High School, this model integrates AI simulation tools (like PhET and custom GPTs) with traditional inquiry methods. The results indicate a significant improvement in students' scientific inquiry competencies (with a 25% increase in experimental design scores) and conceptual understanding (concept inventory scores rose by 18%). The paper details localized case studies such as "AI Analysis of KTX Aerodynamics" and "Modeling the Optimal Angle for Korean Traditional Archery (Gungdo)," providing actionable strategies for frontline physics teachers to bridge the gap between theoretical physics and real-world engineering.

Keywords: High School Physics in Korea; AI-Empowered Modeling; Scientific Inquiry; Process Visualization; 2022 Revised Curriculum