Artificial intelligence adoption and readiness among music teachers and students in Chinese higher education

Abstract Artificial intelligence (AI) is increasingly used for composition, performance assessment, personalized practice, and digital tutoring, yet its acceptance in music education remains insufficiently understood. This study investigates factors associated with AI adoption among music teachers and music-major students in Chinese higher education using an extended Technology Acceptance Model (eTAM) informed by technological pedagogical content knowledge (TPACK). A cross-sectional quantitative survey was conducted across 20 Chinese universities, yielding 300 valid responses, including 150 music teachers and 150 music students. The model examined TPACK, social norms, trust, perceived usefulness (PU), perceived ease of use (PEOU), pedagogical compatibility, creativity concerns, attitude, behavioral intention (BI), and actual usage (AU). Multiple linear regression showed that TPACK significantly predicted PEOU (β = 0.1347, p = 0.033) and was significantly but negatively associated with PU (β = -0.1448, p = 0.014). Social norms and trust did not significantly predict PU or PEOU. PU and PEOU did not significantly predict attitude, while pedagogical compatibility and creativity concerns were also non-significant. Attitude did not significantly predict BI, and BI did not predict AU. These findings suggest that AI adoption in music education may not follow the conventional linear pattern proposed by TAM. Discipline-specific pedagogical values, creative autonomy, institutional conditions, and practical access may be important in explaining the gap between favorable perceptions and actual use.