Auto-Tune was supposed to ruin music. Then it was streaming. Then TikTok. The apocalypse never quite arrives, but the anxiety around each new technology follows a familiar script. Now it is generative AI, and the conversations happening across the music industry sound like the ones that circulated fifteen years ago about digital distribution. Some see opportunity, some see threat, and most are just trying to figure out how to pay rent while the ground shifts beneath them.
The tools have improved faster than anyone expected. What started as glitchy beat generators now produces full arrangements that respond to mood and tempo in ways that genuinely surprise even veteran producers. Bedroom musicians are turning out tracks that would have required a full studio team five years ago. The barriers to entry have collapsed, which sounds like progress until you realize it also means the market is drowning in content. More music exists now than at any point in human history. Standing out has never been harder.
AI as Collaborator, Not Replacement
The generative AI music market hit $0.44 billion last year, and analysts project more than 23% annual growth by 2030. Those numbers reflect integration into everything from Spotify’s personalization algorithms to film and advertising licensing. Thousands of AI-influenced tracks get uploaded to streaming services daily. The sheer volume would be impossible without these tools.
What the numbers miss is context. AI is terrible at knowing what matters. It can iterate endlessly, but it cannot tell you which variation will make someone cry, dance, or remember their first heartbreak. The artists who are thriving treat AI the way earlier generations treated drum machines or samplers, using it to handle the tedious parts so they can focus on decisions that require taste and lived experience. The composition might start with an algorithm, but the choices that shape it into something memorable remain deeply human.
When the cost of experimentation drops, creative courage rises. Musicians are exploring genre crossovers that once felt risky, building immersive sound environments for gaming and virtual reality, and using adaptive music systems that respond to on-screen action in real time. AI has not flattened creativity. It has expanded the studio and given artists access to tools once reserved for those with major-label budgets.
The Rights Problem Nobody Has Solved
Ownership is where things get messy. When an algorithm trains on millions of existing songs to generate something new, who owns the result? The U.S. Copyright Office concluded that AI-generated works can be copyrighted only when they embody meaningful human authorship, meaning music made entirely by AI falls into the public domain. Universal Music Group sued Anthropic for over $3 billion in what could be the largest non-class-action copyright case in U.S. history.
Meanwhile, creators are trying to protect their income in a market where AI-generated tracks can undercut them on price and volume. The old system of tracking royalties through rights organizations was already slow and leaky, and it cannot handle content that multiplies across channels in hours. A song can go viral on TikTok, get licensed for an ad campaign, and spawn dozens of remixes before anyone figures out who should get paid. By the time royalty statements arrive, months have passed, and deductions appear that nobody can explain.
Building for What Comes Next
Some teams are already building infrastructure to embed ownership information directly into music files, using blockchain and smart contracts to automate royalty payments. This ensures that artists are paid the moment their work is used, even as tracks spread across dozens of platforms worldwide. The ownership data travels with the content itself, eliminating the gaps that have allowed value to leak through the system for decades.
Hendrik Hey, a 30-year media veteran and founder of Welt der Wunder, applies this approach through his company, MILC (Media Industry Licensing Content), which creates a permanent, transparent record of ownership and automatically handles revenue distribution.
MILC has extended its infrastructure into music through a partnership with a studio holding more than 70 patents in adaptive and generative music systems, developed in collaboration with Queen Mary University of London. The result is a patent-protected music IP layer that covers film, series, games, and immersive formats, with rights attribution triggered automatically the moment a piece of music is used anywhere within the ecosystem.
A composer whose work appears in a video game released across fourteen countries does not need to track down royalty statements from each territory or wait months for collecting societies to reconcile their records, because the system already knows where the music was used and has already sent the payment. This kind of infrastructure has never existed at scale before, and it addresses the exact problem that AI-generated music is about to make exponentially worse.
The shift away from slow, fragmented rights systems toward real-time, automated infrastructure reflects a broader recognition that the old machinery cannot keep pace with how content moves today. Musicians have survived every technological disruption so far by adapting and demanding better terms, and there is no reason to believe this moment will be any different.
About MILC
Hendrik Hey is the Founder of MILC (Media Industry Licensing Content), a pioneering company in the blockchain and metaverse space, with a strong background in media and content. MILC operates a real live metaverse platform that serves not only the media industry but also various industrial use cases. The company also focuses on Web3 consulting, aiming to support complex real-world industries on their way into Web3. MILC is a sister company of European media giant Welt der Wunder, which Hey founded over 25 years ago. For more information, please visit https://www.milc.global
