As AI moves beyond content generation towards autonomous decision-making and personalised digital experiences, new opportunities are emerging for developers, technology platforms and rights holders.
Artificial intelligence is entering a new phase of evolution, shifting from tools that generate content to intelligent systems capable of adapting existing media, understanding user preferences and making autonomous decisions. As businesses explore the commercial potential of agentic AI, the focus in 2027 is expected to move beyond what AI can create towards how it can transform user experiences, unlock new business models and operate within trusted frameworks for data licensing, transparency and intellectual property protection.
The music industry is emerging as a compelling example of this shift, with advances in AI-powered content adaptation, licensed training datasets and intelligent agents creating opportunities that extend far beyond traditional streaming services. According to Con Raso, Managing Director and Co-Founder of Tuned Global, these developments could fundamentally change how digital content is created, accessed and consumed, while accelerating the need for new approaches to AI governance, commercial licensing and human creativity.
AI will move beyond generating new music into user experience
The conversation around AI in music is already moving beyond the simple question of whether AI can generate a new track. With licensed environments opening up, existing music can be modified, adapted or turned into new versions while the original rights holders, songwriters and artists are still paid.
This year showed the start of the landscape beginning to change, where companies were now starting to successfully license from labels – particularly the majors and large independents. I think what we're going to see in the next year is people starting to really think 'What sort of user experience do I want to allow with agentic AI?'
With a focus on end user experience, I expect the opportunities of AI innovation in music to explode. Moving away from just the concept of creating brand new music, it might be: how do I modify music? How do I create different versions of music? But it's done in concert with a framework that means the original rights holders and songwriters and artists are paid for that inclusion. I think people are incredibly creative, and I believe those journeys are going to skyrocket.
AI disclosure will matter more than an AI-versus-human divide
As AI-assisted music becomes harder to distinguish from conventional production, I think the industry debate will shift toward transparency. Auto tune would have to be one of the first AI tools out there. I would say the same thing with plug-ins and everything that people have traditionally used to create really interesting loops and then modify those loops in different ways. Have we used technology and algorithms before to adapt music for a major artist? Absolutely.
I think it's more about disclosure. If I look at the liner notes and see who contributed to this album, should I know that an AI contributed to that album? If you're going to show everyone else, why not? I think what we'll see in the next 12 months is more structure around how that is shared.
Licenced music for AI training - from bespoke deals to an accessible market
2027 will see the emergence of a more structured commercial market for licensed AI music training, with greater flexibility in how datasets are made available, the rights granted and the commercial terms offered.
The music industry is moving from a period of considerable uncertainty around AI training towards one where developers increasingly want to demonstrate that they are responsible participants, working with rights holders and artists.
I expect the industry to respond with greater availability of properly licensed training datasets, clearer consent mechanisms and more opportunities for artists and songwriters to participate in the commercial benefits.
Like other areas of music licensing, I believe this will evolve beyond individual negotiations towards more flexible commercial models, with different datasets, permitted uses and price points.
The real opportunity is that these models will make licensed music accessible to a much broader range of innovators. That flexibility could become one of the most important catalysts for the next wave of AI music innovation.
2027 could see AI agents become a new consumer of music services
As AI agents become increasingly capable of making decisions and carrying out tasks on behalf of users, the music industry will need to rethink how its technology and data are made available.
We've spent years building music services around how people search, discover and select music. But what happens when it's no longer the person making those decisions? What happens when an AI agent understands what you're trying to achieve and can select the right music for that moment?
This could fundamentally change how music is integrated into other experiences, particularly gaming, fitness, entertainment and personalised digital services.
Today, we can describe music in extraordinary detail. We can identify its characteristics, mood, genre and thousands of other attributes. But an AI agent doesn't necessarily need a list of tags. It needs to understand the music, the context in which it will be used and what it's trying to achieve.
2027 could see the emergence of more music services designed to interact directly with AI agents, allowing them to discover, select and deliver music while respecting the licensing and rights associated with its use.
This opens up a really interesting opportunity. Instead of asking how we build another music application, we should be asking how we make music available to an entirely new generation of intelligent applications. And that could create demand for music in places where a traditional streaming service would never have made sense.