From anthems that ignite our souls, to lullabies that soothe our infancy music
has woven itself into the cloth of our lives, providing solace and sparking
pleasure.
Technological advance has always been one of the major factors in the evolution
of the music world. From the development of instruments to the creation of
digital audio workstations (DAWs), technology has always opened new
possibilities for musical creativity. Today Artificial Intelligence (AI) is one
of the key technologies that have a profound impact on the music industry,
affecting the way music is written, recorded and content is consumed and shared.
The creation of audio AI systems has allowed a wide range of audiences to break
free from any shackles, since now practically a single person can be a composer,
lyricist, and singer, and AI will help them create entire musical compositions
with little to no human involvement[1]. AIVA (Artificial Intelligence Virtual
Artist), Amper Music and MuseNet rely on complicated algorithms to write music
in various styles.
Such AI programs are beneficial to amateurs and professionals
as they allow high-standard music creation without a student having to have a
profound understanding and great training in the field. But the increase in AI
production also brings serious copyright disputes, as well as ethical issues
particularly in the area of copyright[2]. Traditionally, copyright law has been
centred around human creativity and authorship.
The emergence of AI-generated
music challenges these fundamental principles, creating uncertainty regarding
the ownership, originality, and protection of such works under existing
copyright frameworks.
Legal frameworks protect the moral and economic right of composers and
performers; they coordinate 'music below copyright' concerning either the
consumption of the composition or the recording of the composition. Such
possession gives the owner various rights on the ownership of such possessed
resources where they are able to license the resources and get paid in royalty
fees.
It's literally the crime level that protects any person who originated the
song from infringing. As for the conditions that are required for a work to be
patentable, it must be stated that it must be "original" and captured in a
"specific form", as well as the sound of an "appropriate recording on a CD" or
images of "a book made out of sheets of paper saying on it"[3].
It is important
to highlight that musical paints are the sketches of the music and do not
contain lyrics or vocalizations. "Under the scope of Copyright Act[4], it
attains the entire bundle of rights just as any other novel, dramatic or
artistic work"
This makes it slightly more difficult – and more interesting –
than some works such as software copyright. In contrast to other assets,
copyright of the song must not be dealt with as a single work of authorship
under this section; rather, it should be regarded as a collection of various
works, any one of which has different potential copyright protection. This
implies that anyone who contributes to the creation of a track, for example, a
producer may own the copyright to the either of the elements embedded in the
track.
So think of it as a creative pie with pieces cut out. each contributor
can independently claim copyright safety for their piece of the musical pie.
This idea of a "package of rights" permits different creators to be identified
and compensated for their distinct contributions to a track. So, subsequent time
you listen a catchy tune, remember it's a collaboration of multiple creative
minds, every with their own slice of the copyright pie.
Making track via AI has
been trending now, additionally we've got heard of instances of Deep faux videos
of various folks taking the example of deep- faux video of many renowned
personalities like tesla's C.E.O. Allen musk , prime minister Narendra Modi etc.
Functionality of artificial intelligence in contemporary age
Cambridge Dictionary defines AI as a modern technology that simulates human-like
reasoning. The term AI was coined by John McCarthy in 1956, with no universal
definition most scientists maintained their own explanations. Mainly, AI
involves the markdown of language intricacies using vast amounts of digital
data[5].
These can be many books, the entire Wikipedia, and a good sub-dividing
of a billion words harvested from blogs, social media, and just about every
other plausible online platform. The stunning pace of AI development was
motivated by pioneers like Herbert Simon and Allen Newell with the palette of
intelligence by AI changing across intended purposes.
Initially intending to
create general AI, real milestones were achieved beginning in 2010, focussing on
narrow or specialized AI. Generally, enthusiasm for AGI diminished, leading
mainstream scientists to devote their labour towards narrow AI applications[6].
AI And Copyright In Music
About a year and a half ago, Sam Altman, OpenAI's CEO, revealed ChatGPT-a
notable advancement in the AGI restoration. This AI really uses codified
learning. It now really generates responses based on inputs and gradually
refines its skill, hence evolving into what may be termed as intelligence.
Advancements in networks peaked throughout the 1990s and thereafter in the 2000s
as it shifted from CPUs to GPUs.
Presently architectures that have deep
artificial neural networks are overshadowing traditional neural architectures.
This allowed a very impressive lineage of AI upshots to sprout from chess games,
where a predetermined set of rules would govern the moves.
A recent scenario unfolded in the case of Moffatt versus Air Canada. In 2022, an
Air Canada chatbot mistakenly assured a passenger named Jake Moffatt that he
could avail of bereavement airfare. He was informed that he could book a ticket
for his grandmother's funeral and then ask the company to grant him a
bereavement fare. Last Wednesday, a civil resolutions tribunal ruled that when
Moffatt had asked for the fare discount, the airline denied him, saying that the
chatbot had made a mistake and that the request should have been made before the
flight. The airline implied that the chatbot acted independently.
AI music is gaining recent attention; however, its roots can be traced back to
the 1950s. the first example of AI music was made from the program called Lamus,
developed at the "University of Malaga" in Spain. Over the past few years,
AI-composed music has taken the spotlight, as an increasing number of musicians
and composers explore this technology. In 2016, a research team at Sony CSL in
Paris developed an AI system known as Flow Machines.
This program was designed
to help composers craft new music in an existing style or genre, most notably
"Daddy's Car," the first AI-generated pop song that drew inspiration from the
Beatles. Some other notable instances of AI-generated music are "Hello World!,"
a piece by the AI program AIVA, and "The Entertainer," a piano ragtime
composition by the AI software DeepJazz.[7]
Copyright law And the issue Of Music Created With the aid of Artificial
Intelligence
On the matter of copyright law, it deals with cases of injury arising from
infringement of copyright protection and thereby continued surveys contemplation
on whether cases of liability and infringement diminutive in time and space can
adequately be put within the ambit of Copyright law.
To have a clear view on
topics, it is important to have a thorough understanding of "ownership and
authorship" in the context of "copyright law," which serves as a fundamental for
novelty; as chatbots and AI tools like "ChatGPT" quickly come within our reach
and pervade our information economy, we are left clueless about their creation
of creative works and the legal implications that may follow thereon.
"ChatGPT" is an AI language model that generates human-like text in various
contexts; applications include "language translation," "summarisation," and
"question-answering." But the sole use of "ChatGPT" to produce original works
stirs anxiety on the legality of such works. For instance, should "ChatGPT"
generate replies that are considered "defamatory," "discriminatory," or
"otherwise harmful," there may arise legal considerations associated with
liability for such content.
In turn, responsibility for any content created
through "ChatGPT" may now rest with the "individuals or groups" who actually
designed, trained, and deployed the AI model. Moreover, substantial use of "ChatGPT"
would call into question data privacy and the law.
Ownership of AI-Generated Music
Ownership is probably the most contentious aspect in the matter of AI-generated music. In copyright law, normally, it is the human creator, the one who is regarded as the proprietor of the intellectual work. In the case of music created only through some autonomous AI system, determining who the actual owner exists could raise severe questions of legality.
-
Legal Definition of Authorship
Under most copyright perspectives, the one who created a particular work is one who had creative control upon it. Such a definition runs into troubles with AI-generated music, though, where the whole creative process is automated and in which human inputs or intervention may be limited to providing initial stimulus or control parameters for the AI system.
With respect to most jurisdictions across the globe—the USA, European Union, and India—it is not possible to recognise a non-human entity as an author or presenter of copyright. Consequently, the copyright ownership is supposedly impossible to be granted to AI systems and, the questions arise:
- Should the copyright be attributed to the developer that created the underlying intelligent system?
- Or should it be entitled to the user who has provided input and directed the AI?
-
Current Copyright Framework
- United States: The U.S. Copyright Office has so far consistently denied copyright protection to any works produced without human authorship, with such rulings being exemplified in cases dealing with issues related to AI-created content, for instance.
- European Union: The EU Copyright Directive has stressed creativity of humans. Accordingly, AI-generated works mostly are considered not capable in themselves of enjoying any copyright protection.
- India: Indian copyright law which closely mirrors its Western counterparts does not permit copyright on anything but a human author, taking any kind of AI musings therefore within a legal grey area.
-
Case Studies
The myriad legal challenges concerning copyright in AI-generated music have combined into unenviable challenges in the past:
- Amper Music in 2020: A song whose copyright was claimed by the user was allegedly contested against continued ownership claims by Amper Music—the software developers.
- "Daddy's Car": A song made by AI learning the Beatles' music raised the same queries as to the original and authorship of AI-created compositions.
Challenges in Protecting AI-generated Music
Various problems occur with attempts to use conventional copyright principles to AI-generated music:
-
Originality and Creativity
A work must be original and creative enough to fall under copyright. However, most of AI-made music copy from patterns and styles in the existing dataset, raising doubts about their originality.
- Critics contend that AI-generated works are derivative, not original.
- On the other hand, they argue AI systems can present new combinations and arrangements that meet the test of the originality requirement.
-
Absence of Human Authorship
Such being machines devoid of consciousness and creative intent, there can be no normal authorship in AI systems. This leads to one fundamental question: Can a work by a machine ever be original or truly creative, or is it simply a result of programming and data set?
-
Moral and Economic Rights
Apart from copyright, AI-generated music gives rise to moral rights issues, for example, the right to claim one is the author, and economic rights issues, for example, the right to receive incomes from royalties.
- Will the creator or user of an AI system hold moral rights?
- How should royalties be divided when more than one person contributes to an AI creation?
Legal Precedents and Emerging Jurisprudence
There are few laws directly addressing AI-generated music; however, some cases indicate how future courts may handle such issues:
- Naruto vs. Slater (2018): Though not related to music, this case deals with a photograph taken by a monkey. The U.S. courts ruled that non-humans cannot hold any copyright, which might well be a precedent for AI-generated works.
- Thaler v. Commissioner of Patents (2021): This was regarding AI-generated inventions, and the court held that AI systems could not be recognised as inventors under the current patent law. The same logic may apply for copyright.
Future of Copyright in AI-generated Music
Proposed Legal Reforms:
Notably, there are proposals from scholars and industry experts across the board regarding legal reforms and policy changes in the managing of specific issues with AI-generated music. These proposals range from meeting creators, developers, and users halfway to promoting innovation in the music industry.
-
Introducing AI as a Co-Creator
One suggested remedy is that of recognising AI as an accomplice in creation with humans. Here would be the terms of this model:
- Dual ownership of some percentage of copyright will be awarded to the developers of AI, as they are providers of the underlying technology and algorithms.
- It would award some patent rights to users who direct AI by input content or selection of parameters to generate them.
This provides acknowledgment of collaborative creativity in AI-generated music while ensuring that both developer and user will demand some part of the final product.
-
Introduction of a New Type of Copyright
Another remedy suggested is to create a new class of copyright tailored for the works of AI. This would have provisions regarding:
- Ownership: assigning which may be the active party in awarding the copyright — either to the developer, user, or both.
- Duration: limiting it to ensure that copyrights do not last long to incentivise innovation and public ease.
- Licensing: determining a licensing and a royalty distribution system.
-
Royalty and Royal Sharing Models Establishment
The royalty-sharing model could be structured to distribute revenue from AI-generated music among all categories of stakeholders:
- Developers
- Users
- Traditional artists whose works serve as training input
Thus, this model will ensure fair compensation for every party involved without letting concerns arise over the devaluation of human creativity.
-
Integration of Blockchain and AI Copyright Management
Blockchain holds the promise for settling almost all copyright controversies under which AI-generated music would fall. By providing a decentralized open ledger, it could:
- Track Ownership and Attribution Record: Something like the contributions of developers, users, and AI systems ensures the accuracy of ownership and attribution.
- License and Royalty Management: Automate the licensing process and disburse royalties through smart contracts, significantly reducing administrative costs while ensuring timely payments.
- Increased Copyright Protection: Registry proof against tampering makes it more convenient for copyright claims enforcement as well as dispute resolution.
Many companies and organisations are already taking steps toward blockchain-based copyright management in music, pointing toward a possible future of decentralised copyright systems.
Ethical and Policy Considerations
Legal reforms alone will not suffice; ethical and policy dimensions should also be addressed to ensure responsible use of AI in music creation.
-
Innovation versus Creativity
Policies should aim to resolve the conflict between fostering innovative technology and protecting the rights of human creators. While AI-generated music opens new horizons for creativity, it should not devalue human artistry or discourage traditional musicians.
-
Compensating Fairly
It is vital that human artists, developers, and other stakeholders are properly compensated as AI systems grow in influence. This includes paying original artists whose works are used as training data for AI systems.
-
Addressing Cultural and Societal Impacts
AI-generated music affects societal and cultural understandings of creative acts and ownership. Policymakers need to consider how AI impacts culture and should support diversity, inclusion, and cultural preservation within the music industry.
Conclusion
Artificial intelligence has brought the music industry to the threshold of its
own opportunities and conflicts with copyright law. While it has democratised
creative effort and broadened the scope for creativity, it has raised complex,
multi-faceted questions of legal, ethical, and economic concern over authorship,
ownership, and originality.
To meet these challenges, it is important to
Legalise frameworks that provide for the collaborative nature of AI-generated
music lImplement more technological solutions such as blockchain for better
copyright management , Balancing interest between creators, developers, and
users without inhibiting innovation while protecting human creativity. As AI
continues to evolve, so will the legal and regulatory landscape around it to
make music remain vibrant for both machines and humans.
End Notes:
- John McCarthy, Artificial Intelligence and Music: A New Frontier, Journal of Music and Technology (Oxford University Press, 2023).
- AIVA: Artificial Intelligence Virtual Artist, AIVA Technologies (AIVA, 2024), available at: http://www.aiva.ai (last visited on December 10, 2024).
- David Vaver, Intellectual Property Law (Oxford University Press, London, 2021).
- Copyright Act, 1957 (No. 14 of 1957)
- John McCarthy, "Artificial Intelligence: The Scientific Revolution 27 AI Magazine (Association for the Advancement of Artificial Intelligence, 2006) 12–15.
- Narrow AI and AGI: A Historical Perspective, AI and the Future Journal, available at: http://www.aifuturejournal.com (last visited on December 10, 2024).
- AI-Generated Music: The Complexities of Copyright Protection" (Legis Music – Music Licensing for your Videos & Business), available at: https://legismusic.com/copyright-protection-ia-music/ (last visited on February 16, 2024).
- 244 USA (2018)
- 322 USA (2021)
Comments