One inevitable question arising from the progress of artificial intelligence in creative domains, such as music composition, is a substantial legal network. AI-generated music, of which platforms like Suno AI are exemplars, has also begun threatening accepted thoughts on copyright and ownership in the music world.
This article dives into the tangled legal web of AI-made music, as we look at where issues with copyright and ownership overlap and eventually what this may mean for artists, developers of AIs or potentially everyone in the wider world of music.
Understanding Copyright in Music
What does it mean that AI creates music First of all, before we can start discussing the details about how music generated by an algorithm functions in practice, let us first see a bit about{ what are copyrights according to law}:
Musical Work: It refers to the composition that underlies a recording or performance.
Sound Recording – This protects the recording of a particular musical composition.
Copyright law generally provides these rights to the creators of works which are humans. What happens, however, when the music is created by means of a machine?
The AI Challenge to Traditional Copyright
AI-generated music presents several challenges to traditional copyright law:
1. Authorship and Originality
Copyright law typically requires a human author and a degree of creativity or originality. AI-generated music raises questions about whether a non-human entity can be considered an “author” in the legal sense.
2. Ownership
If AI can’t be an author under current law, who owns the copyright to AI-generated music? Potential claimants might include:
- The AI developer
- The user of the AI system
- The owner of the AI system
- The trainer of the AI model
3. Derivative Works
AI models are trained on existing music. This raises questions about whether AI-generated music could be considered a derivative work of its training data, potentially infringing on existing copyrights.
4. Public Domain and Fair Use
There are ongoing debates about whether using copyrighted music to train AI models falls under fair use or requires licensing agreements.
Current Legal Approaches
Current Legal Approaches
In 2024, there is no accepted worldwide response as to how the copyright of AI-generated music should be handled. Nevertheless, a few trends and strategies are starting to appear:
1. Human-in-the-Loop Approach
In such circumstances, governments may opt for a “human in the loop” regime under which copyright protection is bestowed when AI was only used as an instrument and there were substantial human efforts or curation involved with respect to output. This could include:
- Setting parameters for the AI
- Selecting and arranging AI-generated elements
- Post-processing or editing the AI output
2. Work-for-Hire Doctrine
Instead, AI might be treated as more of an instrument – akin to musical instruments or software. This approach would result in copyright being vested with the individual or entity using an AI since it happens today for every other tool used as a part of creation.
3. New Categories of Protection
Nonetheless, there have been calls for the development of new modes of intellectual property protection that address AI-generated works. For instance, grants of protection may be shorter or give rise to different sets of rights than that traditionally associated with copyright.
4. Public Domain Approach
Others say that works generated by AI should automatically be in the public domain since there is no individual author for it and without an active level of creativity needed to qualify for copyright protection.
Real-World Cases and Precedents
Although there have not been many court cases quite specific to AI-generated music, some related cases provide indicators of how courts may deal in relation with this problem:
Naruto v. Slater (2018): This human was not about AI, but the case established that non-human entities cannot hold copyrights under U.S. law as seen in this “Monkey Selfie” case is a well-established principle under copyright policy and laws since long before Naruto decided to stand trial over his selfie title feud for mankind.
Feist Publications v. Rural Telephone Service (1991): This U.S. Supreme Court case holds that there is a “minimal degree of creativity” required to grant copyright, which could establish something of an AI-generated work threshold
Acohs Pty Ltd v Ucorp 34 Pty Ltd (2012): An Australian case focusing on works that were created by a computer, stating protection would arise where there was enough human input.
Implications for the Music Industry
The legal grey area on AI-created music is a significant barrier to the industry:
1. Licensing and Royalties
Licensing such music and distributing royalties might be complicated by ownership uncertainty.
2. Collaborations with AI
Creating rights to protect music resulting from collaborations between musicians and AI systems
3. Music streaming services and music libraries
This will entail specific changes on the part of these platforms to prove copyright ownership in AI-generated works.
4. AI Training Data
The music industry may have to introduce guidelines or licensing frameworks around how copyrighted material can be used as AI training data.
Potential Solutions and Future Directions
Outlined here by JRL are several potential solutions, as the technology continues to evolve.
1. Legislative Updates
Various experts have suggested that copyright laws should be modernized to more specifically acknowledge the implications of how AI forms its works. For example, this could mean inventing entirely new forms of protection or specifying the copyright conditions imposed on works created by AI.
2. Blockchain & Smart Contracts
With blockchain technology, it may be possible to track the provenance of AI-created music and automate licensing and royalty payments.
3. Ethical AI Development
This could also address some of the legal concerns by developing ethical guidelines for most AI music generation scenarios (making it transparent that there is an involvement from AIs and introducing a fair compensation model).
4. International Harmonization
The music industry is global and international collaboration will be necessary to lay down consistent rules for AI-created tracks worldwide.
Conclusion
The legal field of AI-generated music is grey and evolving xmlDoc Block ohai 9sec. AI is advancing faster than our comprehension of what it means for creativity, authorship, and ownership in music. Recent legislation, important court decisions, and the actions of established industry players will likely bring significant changes to how these issues are dealt with in the medium term.
At this point, any musicians or AI developers entering the music industry should not disregard the legal uncertainties at hand for their songs created by artificial technology. All in all, it is important to treat the adoption of AI technology for music creation with due care and attention – when in doubt seek legal advice and keep an eye on what happens next.
As this new territory develops, it will be critical to ensure that we balance encouraging innovation in AI music tech with the need to protect human creator rights and motivations. The solutions we design will build the landscape that thousands of producers in future years will create music, changing the creative options they have at their disposal but also reshaping how authorship is realized digitally.
The legal issues of AI-generated music are more than just ivory tower speculation but rather critically important for the entire law stack related to the music industry. How we answer these questions will significantly influence the shape of music making, distribution, and consumption carried out by people vs. robots in 2018 – but I invite you to tune back into AIesthetics next year for more on developing practices!
