As debates about the implications of artificial intelligence (AI) for the future of music continue, the role of AI in music and music education remains uncertain (Herington et al., 2025). To address this issue, in 2025, our interdisciplinary research team conducted a national survey of musicians to gather information about:-Their experiences using AI for music and non-music purposes-Their perceptions about using AI in music-making and music teaching and learning-How AI could improve music educators’ music teaching and learningOut of the 226 respondents, 126 self-identified as music educators. The music educators' responses were broken into three groups based on who had not experimented with AI either within or outside of music (Group A, n = 39), had used AI outside of music, but not for music-making (Group B, n = 48), or had used AI for music-making (Group C, n = 39). Our findings encompass data on usage, interests, risks, and benefits. We analyzed the survey data using descriptive and inferential statistics, including analysis of variance with Bonferroni post-hoc tests to examine differences in means among these three groups. We found that the music educators in Group C demonstrated a significantly greater interest in using AI for co-composing than those in Groups A and B. Group C also had a significantly greater interest than the inexperienced AI group (Group A) in using AI for source separation, sample generation, effect plug-ins, and instrument plug-ins. Additionally, when considering the opportunity and risk of using AI for their own music-making, the experienced AI music groups’ view of risk was significantly lower, but their view of opportunity was significantly higher, versus the inexperienced AI groups’ view of risk and opportunity. All three groups expressed ethical concerns about the use of AI in music.The music educators were also asked an open-ended question about how a magic wand could assist them in their music-making. Using axial coding (Saldaña, 2021), the music educators’ responses (n = 78) were organized into the following categories, depending on whether their wish involved workflow, creativity, musicianship, accessibility, clerical tasks, and time. Answers to this question revealed specific aspects of music teaching and learning or technological annoyances that could potentially be resolved with the assistance of AI.These findings suggest a connection between music educators’ AI experience and their perceptions of AI’s impact on music, calling for more personal experiences with AI as musicians determine its usefulness.