,
By Mark Frieser
CEO, Sync Summit
I’ve been doing some deep thinking on the potential impact of AI-generated music on all of us who work the field of music for visual and interactive media, and today I’d like to share my thoughts with you on how AI-generated music is poised to reshape the way music is made, integrated and used for visual and interactive media, and how rights holders and artists can be protected and even empowered in this evolving creating/business landscape.
But before I do that, I’d like to address the divide between the content and tech industries when it comes to the importance of respecting and protecting intellectual property.
I’ve been involved in tech and content a long time, and the one constant I’ve witnessed is that tech developers and entrepreneurs will move fast and break things to bring new solutions to market as quickly as possible, IP considerations and existing market models be damned. This is especially true at the initial stages of company and market development, leaving IP owners attempting to enforce and protect their property rights after the tech has been launched and adopted by the mass market.
This happened when file sharing platforms like Napster, Kazza, Limewire and a plethora of other file sharing platforms launched at the turn of the 21st century. It happened again when a seemingly endless phalanx of ringtone and mobile content companies launched in the early 2000s. It happened at the advent of streaming networks. And it’s happening now with the growth and adoption of AI-generated music platforms, most notablty Suno and Udio, both of which use unlicensed musical IP to train their AI.
The point is, the cavalier disregard shown by companies like Suno and Udio toward intellectual property law is nothing new in the tech. This mindset is hardwired into the modus operandi of many technologists and executives. And if we were to take some of them at their word, they believe IP law should be abolished altogether.
Case in point: On April 11th 2025, Twitter founder and Block CEO Jack Dorsey threw down a gauntlet when he Tweeted “Delete All IP Law”, which of course led Elon Musk to reply “I agree”.
Statements like these coming from two founders of multiple publicly traded companies whose fortunes rely on IP ownership, proprietary tech, patents and copyrights, are not only irresponsible, asinine and disingenuous, they’re deeply flawed thinking.
IP protection doesn’t stifle innovation. It fuels it. It shouldn’t be dismantled or deprioritised in favor of open systems. Arguing that IP should be scrapped may sound bold in an era of fast-moving tech and democratized creativity, but it completely overlooks the foundational role IP plays in incentivizing, protecting, and rewarding creative work – especially in music.
IP protection isn’t a barrier to creativity; it’s a catalyst. It allows artists, composers, producers, and music companies to own the fruits of their labor, negotiate fair deals, and reinvest in new work. Without it, the economic foundation of creativity falls apart.
AI, with all its promise, must not become a vehicle for exploitation. Specifically within the music-for-media sector, ethical IP usage to train AI is essential if we’re to harness the potential of AI-generated music responsibly.
Training With Permission: A New, Ethical Large Music Model (LMM)
As mentioned earlier, most first wave AI-generated music platforms, most notably Suno and Udio, while technically impressive, trained their models on music they had no permission or consent to use.
This makes their output, in its current form, completely unusable for visual and interactive media production.
Why? Because music generated from models trained on unlicensed IP exposes everyone in a production—brands, networks, studios, producers—to legal risk, higher, unexpected costs, production delays and in extreme cases the withdrawal of content or ads from distribution networks after its release.
In short, if AI-generated music is to be used in visual/interactive media, verifiable IP permissions and ownership are essential. And on this crucial point, today’s leading AI music generation companies fall short.
That said, big changes are afoot. AI-generated music, and how it is created, is evolving rapidly. This evolution will allow AI-generated music to fulfil a market need in the music in media market. Within the next year, new players will emerge that will deliver high-quality AI-generated music from systems trained ethically with verifiable IP standards that meet the requirements of everyone in the visual/interactive ecosystem.
And, it is this second wave of AI-generated music companies training their Large Music Models (LMMs) exclusively with licensed/owned content who will deliver on the promise of AI-generated music for visual/interactive media..
These ethically trained LMMs will use music that was either composed specifically for training the LMM or properly licensed from rights owners for that purpose. This will ensure transparency and ethical alignment with copyright law. This significant shift is not just technological, it will change how media and music industries perceive and use AI-generated music, and will highlight in a very practical way how the tech industry can be a trusted partner of music industry, and more broadly the content production industry, rather than an adversary.
A Major Opportunity for the Entire Music for media Ecosystem
AI-generated music platforms, when executed ethically, can provide powerful benefits to visual/interactive media content owners, producers and creators while at the same time ensuring the integrity and monetisation of music copyright, music rights holders and creators.
Let’s take a look at some of the potential benefits for creators in visual/interactive media:
- Streamlined Workflows: Music can now be generated, edited, and integrated within minutes, drastically reducing turnaround times in both pre-and post-production.
- Cost Savings: Custom scoring has traditionally been expensive and time-consuming. AI can lower licensing costs, reduce revision cycles, and optimize budgets.
- Hyper-Customization: AI can empower producers and creatives to quickly and efficiently customize a song within minutes to create a orchestral mix, a lo-fi mix, or a trailer mix edited to the specific purpose, feeling and emotion of the usage.
- Scalable Creativity: AI enables editors, creatives, music supervisors, studios, agencies and brands to ideate at scale, auditioning countless musical directions before selecting the perfect one.
Most significant, when used correctly, generative AI will greatly benefit the content industry as a while. A 2023 Deloitte report found that generative AI has the potential to reduce content production costs by up to 40%, a figure likely to grow as AI-generated music becomes more embedded into production pipelines.
These are just some of the advantages that point to why AI-generated music is set to become an essential part of how we create and use music for visual and interactive media. The key question is: how do we harness these new paradigms and tools while protecting—and empowering—creators and rights holders?
A Critical Juncture for AI-Generated Music and the Music Industry in General
While the upside of AI-generated music is undeniable, it must not come at the expense of musicians and rights owners. AI should not be used as a way to dodge licensing fees or devalue existing music catalogs.
Training AI on copyrighted work without consent is like taking someone’s land, tools, and materials to build your house—then refusing to pay. It’s theft dressed up as disruption.
To balance innovation and rights protection, we need robust IP protection that takes into account the realities of technical development, evolution and innovation. This requires agreements on ethical standards and the development of AI training models that ensure artists and rights holders are compensated when their work contributes to AI-generated content—even indirectly through training data.
While the current IP laws and structure allow us to build and scale services that create AI-generated music ethically, they are only the starting point.
We can and must ensure IP adapts to market and technical evolution and there is a remuneration structure that rewards rights holders whose works trained the LMM. To do this, technologists, rights owners and regulators must partner to create ethical licensing frameworks that allow music to be used for training while still honouring creators’ rights and value, providing revenue to music IP owners fairly and on an ongoing basis.
Opportunities AND CHALLENGES for Music Creators and Rights Holders
So with all of this said, where are the benefits of all this for music creators and rights holders? Overall, these new services and platforms will be a net positive for the music in media industry, opening up new models for revenue and growth for music companies and creators. Here’s three examples:
- New Revenue Models: Musicians and rights holders can opt in to license their work for AI model training, establishing long-tail value streams as their sounds become the building blocks of new creations.
- AI-Enhanced Creativity: Composers can (and many already do) use AI as a co-writer, quickly iterating ideas and variations, saving time for high-value creative work.
- Tool Not Threat: Like drum machines and synthesizers before it, AI won’t replace composers and artists—it’ll amplify the ones who adapt.
This said, it would be disingenuous to say this market and technical evolution will not present significant challenges to many players in the current music for media ecosystem. Specifically, music libraries and custom music houses will likely need to pivot at least part of their value proposition to survive and thrive in the evolving market, offering AI-enhanced catalogs, dynamic creative tools, or hybrid services that combine human and machine creativity alongside current service offerings.
Charting a Positive Path Forward
Whether your’e someone who is a change agent, embraces change, fears it or fights against it doesn’t matter because the hard fact is that a significant portion of the music that is used or created for visual and interactive media is going to be made with AI. What we need to focus on is how we can work together to ethically harness its potential.
By training LMMs with ethically sourced, licensed, commissioned and paid for music, AI-generated music platforms can provide the visual and interactive media industries with more productive, more scalable, more customisable and more cost-effective solutions that meet the legal and creative needs of the media industry while prioritising transparency, fair licensing, and creative collaboration.
Ultimately, AI-generated music is not the end of music creation—it’s the next chapter.
With strong IP protections, ethical training, and open collaboration, we can create a future where music is more customizable, more accessible—and still deeply human.
