A New Approach to AI Music’s Copyright Challenge

An independent perspective from Dutch music-rights specialist Marvin Kuijs on how AI could make music rights more transparent and fairer.

AMSTERDAM, NETHERLANDS, October 27, 2026 /EINPresswire.com/ -- What if AI could do more than create music? What if it could also reveal who contributed to it?

A System That Already Exists

The debate around AI-generated music usually focuses on one question: can it be protected by copyright? A more interesting question is whether AI could help distribute music rights more fairly and transparently. The existing system used by performing rights organisations (PROs) such as Buma/Stemra in the Netherlands provides a useful starting point for considering how this might work. For a song with lyrics, rights are generally divided among three roles:

- Composer: 41.67%
- Lyricist: 41.67%
- Arranger: 16.66%

Now imagine that a songwriter creates the melody and lyrics but uses an AI platform such as Suno to generate the instrumentation, rhythm and production. In effect, AI performs the arranger’s role. Under the existing model, the composer and lyricist would retain their shares, while the arranger’s share would require further consideration. This raises an important question: if AI creates the arrangement, who should receive that share?

The Missing Link

AI models do not create music in isolation. They are trained on enormous amounts of existing music. Much of today’s debate concerns whether that training should be permitted, but less attention is given to what AI could do with this information. Imagine an AI platform producing not only an arrangement, but also a transparent report explaining how it was created. The report might show that an arrangement was influenced equally by two existing musical works.

Instead of allocating the entire arranger’s share to the AI platform, that 16.66% could be distributed among the rights holders whose music contributed to the result. If two works contributed equally, each could receive 8.33%. If twenty works contributed, their rights holders could collectively share the same 16.66%. The underlying system would remain unchanged. Only the distribution of the arranger’s share would become more transparent.

The Infrastructure Already Exists

The music industry would not necessarily need to build an entirely new royalty system. Collective management organisations already collect royalties, maintain ownership records and distribute income to composers, lyricists, publishers and other rights holders. Under this model, they would simply receive more detailed information. Alongside the composer and lyricist, an AI attribution report could explain how the arrangement was created and how the arranger’s share should be allocated.

If an original work had multiple composers or a publisher, the organisation would already have the ownership information needed to distribute the relevant royalties correctly. The major change would therefore be required from AI platforms, not rights organisations. They would need to become more transparent about how their music was created.

A Second Example

Consider a different scenario: a person writes only the lyrics, while AI generates both the composition and the arrangement.
Copyright law is still developing in relation to such works. However, a transparent attribution system could identify which existing musical works influenced the composition and arrangement, and to what extent.

The human lyricist would retain their share. The remaining rights could be distributed among the original rights holders whose work demonstrably contributed to the AI-generated composition and arrangement.
Nobody would lose their existing rights. Instead, everyone who contributed to the final work could receive a proportionate share.

The Next Step for AI Platforms

If AI platforms documented this information, they would become more than music-generation tools. They would also contribute to the administration of music rights. Every AI-generated work could be accompanied by an attribution report explaining which existing works influenced the composition, the percentage attributed to each, and how the relevant shares could be distributed. That information could then be submitted to the appropriate rights organisation.

This would give the music industry something it has never truly had: greater transparency about the origins of AI-generated music.
The discussion around AI often focuses on who might lose. But AI could also help ensure that everyone who contributes to a musical work is recognised and rewarded more fairly.

Perhaps AI Isn’t the Problem, but the Solution

Today, the debate over AI in the music industry is largely being fought in courtrooms. But perhaps it's time to ask a different question. Not how we can stop AI-generated music, but how we can use AI to make the management and distribution of music rights more transparent and fairer.

Marvin Kuijs says: “Instead of spending years fighting each other in court, AI platforms, publishers and PROs could work together to explore how we can build a transparent system that ultimately benefits everyone. If the music industry ever decides to take that step, I would be happy to contribute more than 25 years of experience in copyright, music publishing and artist development to help create a practical and workable framework.”
Perhaps artificial intelligence's greatest contribution will ultimately not be its ability to create music, but its potential to help us build the fairest and most transparent music rights system the industry has ever known.

Marvin Kuijs
musiQproof