The music industry has already moved beyond the issue of whether artificial intelligence will affect music to the more complex one of how. I argued in a separate article that the urgent question is: What kind of music technology ecosystem will emerge that uses AI and remains fit-for-purpose for the industry?
To answer that top-line question, it is first necessary to address nine other questions regarding how and where AI will have an impact throughout the industry.
- What does "AI Music" mean? We need to distinguish production tools, full-generation platforms, fan products, sync licensing tools, and voice or sound-alike products; separate administrative assistance from creative substitution; and distinguish music as an end-product from music as part of a greater creative endeavor. The "category error" is to treat all of these as a single, indivisible, moral, and legal object. Each aspect of AI use will require distinct industry inputs and responses.
- What does "Musicianship" mean? Music-making is intimately wrapped up in the evolution of an artist's own identity rather than being mere content to be generated and marketed. Some friction in the creative process is wasteful – dealing with software and hardware configurations, managing assets, etc. – whereas other friction is inherent and arguably necessary – writing, choosing, failing, revising, and finally recognizing that something works.
- Can the industry move from litigation to deals? Litigation remains a necessary deterrent to bad actors, and legislation may be needed where existing law is fragmented, especially around voice, image, name, and likeness (VINL). The industry is already responding with new licensing architectures and needs to complement these with guardrails and rules governing, for example, what protected content can be used for AI training purposes, or what an AI system can generate when invoking a specific artist, recording, composition, voice, or style. Artist consent may well become the non-negotiable factor, and a healthy market must embrace that.
- What are the future economics of creative work? Different models of compensation will continue, from one-time buyouts to revenue participation that grows as services grow, upfront payments, ongoing or fractional royalties, equity participation, and remuneration for contributions to model training. The unresolved problem is attribution: if AI used to generated output is influenced by many works, how should payment be allocated? This is not just about financial compensation – it is increasingly about transparency and trust in the ecosystem.
- Is the existing music tech infrastructure "fit-for-purpose"? Effective industry governance depends on ecosystem infrastructure. The music technology ecosystem must connect creators, creations, distribution, detection, licensing, monetization, recommendation, and listener-facing disclosure. As AI-assisted creativity produces new content as well as new tools and processes, music production tools will need to become flexible hosts for workflows rather than standalone closed systems. Interoperability is not a technical luxury.
- How does the industry respond to oversupply and fraud? An A&R professional used to be able to hear nearly everything released in a genre. The prevalence of DAWs, home recording, cheap distribution, and streaming already undermined the old scarcity model, but AI has made that abundance cheaper, faster, and easier to exploit. AI-generated tracks can be created at scale and uploaded under fake identities, then streamed through bots or click farms. There is a wholesale collapse of scarcity, and as streaming royalties are often pooled, fraudulent streams create economic leakage from real artists to bad actors.
- Can product design and workflow reinforce trust? Users are asking tougher questions about AI training data, rights, commercial safety, and platform rules. Music tech companies need to be transparent about what their tools do, what data they use, what rights are granted, and what risks remain. Where the traditional licensing system is too slow, costly, or confusing, AI tools that offer "rights-clear" outputs become more attractive. Creators are not rejecting human creativity; they want the tools they use to match their own workflow. Trust ought to be a product feature.
- Are new models, such as "fan co-creation", consistent with artists' own interests? Fan tools can let users move from passive listening to active creation within an artist's own world and brand but as a distinct, licensed category. Fan co-creation reframes AI use from replacement to new opportunities and new relationships with an artist. Licensed properly, this can create new revenue, deeper engagement, and new forms of participation. The legal and product architectures will need to reflect those differences.
- What is the future of music education? "AI literacy" should be embraced within music studies to reduce fear and learn what is useful and help students understand what AI systems can and cannot do, where they help, and where they deaden human creativity. "Rights literacy" should help students understand AI training and outputs, implications for VINL, authorship, contracts, and platform rules. "Infrastructure literacy" should cover metadata, attribution, distribution, fraud, labeling, and disclosure. Finally, "artistic literacy" should emphasize preserving taste, judgment, personal processes, and individual expression in the face of more ubiquitous use of AI.
As organizations navigate this evolving landscape, they should continue to monitor legal, regulatory, and commercial developments that will shape the future of AI in music. If you have questions about these issues or how they may affect your business, contact author Peter Brown for more information.