AI Music Adoption Spreads, but Usage Varies Widely Across the Industry
A growing wave of musicians are turning to artificial‑intelligence tools, yet how deeply they embed AI in their craft differs sharply between tech‑centric circles and the wider creative community.
A 2025 survey by LANDR—a cloud‑based production platform—found that 87 % of its 1,241 respondents tapped AI somewhere in their workflow, and 69 % had increased their use compared to the prior year. Moises, an AI‑powered music‑editing app, reported that 67 % of its 1,525 users had engaged AI in music‑related tasks over the past year. Within that group, 78 % of professionals and 60 % of hobbyists said they used AI. Those figures contrast with broader rights‑society surveys, where adoption ranges from 29 % to 47 %.
The gap stems partly from who is surveyed and how “AI use” is defined. LANDR and Moises recruit participants from platforms that already host AI tools, creating a self‑selecting sample more likely to adopt such technology. Rights‑society studies—APRA AMCOS (38 % of 4,274 members in Australia and New Zealand), Teosto (47 % of 1,108 respondents in Finland), and SGAE (34 % of 1,257 members in Spain)—sample a broader cross‑section of the industry and focus narrowly on generative composition.
Beyond raw numbers, the surveys differentiate how AI is employed. Some musicians use it for tasks that require little creative input—stem separation, restoration, or mastering assistance. Others integrate AI into a hybrid workflow: they write a full song, then direct AI to generate hundreds of arrangements, swap stems, overdub new performances, and finally mix the track themselves. A third group relies on AI to produce entire tracks from a brief prompt, often generating large batches with minimal human review. The most extreme cases involve automated “track factories” that churn out thousands of songs, assign metadata, and upload them to streaming services with little oversight.
These practices carry varied industry implications. AI‑assisted production can lower technical barriers, allowing songwriters with limited resources to hear full arrangements quickly. Yet the ease of mass‑producing music raises concerns about market saturation, potential copyright infringement, and the dilution of royalties for human performers and producers. The Australian government’s 2025 decision to reject a broad text‑and‑data‑mining exception—which would have allowed AI developers to use copyrighted works without permission—highlights the legal uncertainty surrounding AI‑trained models. In July 2026, the government reaffirmed that creators retain ownership and control over their works, but practical enforcement mechanisms remain under development.
Stakeholders are also wrestling with attribution and disclosure. A 2024 letter from the Artist Rights Alliance, signed by more than 200 artists, called for an end to predatory AI practices that imitate identities or replace human work, while acknowledging that responsible AI can support creativity. The letter clarified that using AI for tasks such as stem separation or restoration does not amount to claiming authorship of a finished track.
In short, AI tools are increasingly embedded in music workflows, but the degree of human involvement varies widely. Adoption rates differ across communities, and the legal framework remains unsettled. As the technology matures, the industry will need to refine definitions, establish clear attribution practices, and develop robust licensing mechanisms to balance innovation with creators’ rights.
