If it’s not measured, is it happening at all?

A recent scoping review in Research Studies in Music Education (Iversen, 2026) synthesized 22 peer-reviewed studies on AI in music classrooms published between 2023 and 2025. The findings are encouraging: gains in student motivation, self-efficacy, technical skill, and several dimensions of creativity, particularly when AI is paired with teacher guidance rather than used to replace it.

Look closer at where that research came from, and a gap opens up. Seventeen of the twenty-two studies came out of a single country. That means the vast majority of the world’s music educators, across dozens of educational systems, cultures, and teaching traditions, are essentially absent from this body of research. Only two studies touched voice or choir at all, compared to four on piano alone. The review’s window closes in March 2025, and its author is candid that the search covered English and Scandinavian-language academic databases only, noting that classroom-level experimentation elsewhere may well be happening without yet appearing in that literature.

We think that gap is real, and we think we know part of why it exists.

Our own research into AI adoption among choral educators, presented this year at the Texas Choral Directors Association Convention, asked what predicts whether a teacher intends to use AI at all. The strongest predictor, by a wide margin, was Social Influence: peer modeling, not administrative mandates or published outcomes data. Teachers pick up a new tool because someone they trust, in a room they respect, is already using it and talking about it.

That has a direct implication for the research record. If adoption spreads through relationships rather than through published evidence, classroom use can accelerate well before it shows up in a database that peer review takes a year or two to populate. The growth we’re watching in real time supports this: TMEA’s AI programming grew from one session and roughly 80 attendees in 2025 to six sessions and an estimated 500 attendees in 2026. We’ve also seen a noticeable uptick in posts and conversation following this year’s ISME conference. None of that shows up in a citation count. It shows up as rooms filling up.

Our TCDA research also surfaced barriers that don’t appear in the broader literature at all: concerns about AI’s threat to artistic identity, and its environmental cost. Those aren’t the kind of concerns a study resolves. They’re the kind that get worked through in conversation with someone who’s already sitting with the same question, which is exactly the mechanism Social Influence describes.

Put together, our read is this: music educators across most of the world aren’t missing from the research because nothing is happening in their classrooms. They’re missing because what’s happening is spreading peer to peer, faster than the publication pipeline can capture it, in a discipline the existing research barely touches.

Peer-led professional development is the piece we think matters most here, and we want to be clear that this shouldn’t be a responsibility placed on individual teachers alone. It requires the system around them to support it: dedicated time to engage with current research and practice, backing from schools and competitions that actively encourage AI-facilitated pedagogy, and teacher communities structured to sustain collective learning rather than leaving each educator to figure it out solo.

MusEdLab sees this gap and intends to help close it. That means investing in peer-led PD, building learning content built around how music educators actually work, developing tools designed specifically for music education rather than adapted from general ed-tech, and meeting teachers where they already are instead of asking them to come to us.

The research proving AI can work in music classrooms is in good shape. What’s still being built is the infrastructure that helps music educators learn from each other. That’s the part worth investing in next.

Sources: Iversen, K. T. (2026). Impacts of Artificial Intelligence in Music Education: A Scoping Review of Instructional Strategies and Student Learning Outcomes. Research Studies in Music Education. MusEdLab.ai original research, “AI Adoption in Choral Music Education,” presented at the TCDA Annual Convention, 2026.