This comprehensive examination delves into the intricate confluence of
artificial intelligence (AI) and copyright law. It investigates whether
AI-generated works can be granted ownership and copyright protection,
scrutinizing issues of originality, authorship, liabilities in cases of
infringement, and the applicability of fair use. The analysis reveals that
AI-generated works often lack the originality required for copyright protection.
Authorship questions arise due to AI's absence of human consciousness.
Determining liabilities in copyright infringement cases involving AI remains
complex, as does deciding the duration of copyright protection for AI-generated
works. The study also explores the doctrine of fair use in the context of
AI-generated content. In conclusion, it suggests the need for legal clarity,
international collaboration, fair use evolution, responsible AI development, and
efficient dispute resolution mechanisms to balance the rights of human creators
with AI's unique role in content creation.
Introduction:
The intersection of artificial intelligence (AI) and copyright law has brought
forth a myriad of intricate legal questions and challenges. Central to this
confluence is the fundamental query of whether AI-generated works can be endowed
with ownership and protected under copyright law which is essential for the
protection of the faith of the copyright owner as it is an essential element for
determination of the liabilities in case of infringement of the work which has
been granted copyright protection.
This inquiry unravels a multifaceted
exploration into the very essence of originality and creativity in works crafted
by AI, the vexing question of authorship in the realm of AI-generated content,
and the intricate determination of liabilities in cases of copyright
infringement involving AI creations.
This discussion embarks on an in-depth
examination of the multifaceted dimensions of this complex issue, scrutinizing
both the legal frameworks and the practical implications of AI-generated works
within the context of copyright law.
It grapples with a series of critical
questions, including whether AI-generated works genuinely exhibit the required
level of originality as mandated by copyright law, or if they predominantly
derive from learned patterns and data-driven processes. Furthermore, the
discourse delves into the contentious issue of authorship within AI-generated
works, given the absence of human consciousness and subjective creative intent
in AI.
It seeks to address questions about who should be designated as the
author of these works and how such authorship status influences copyright
ownership. In cases of copyright infringement involving AI-generated works, this
exploration endeavours to untangle the complex web of determining liabilities.
It probes into the question of who should bear the responsibility for copyright
violations linked to AI creations.
Lastly, the discourse explores the relevance
of the doctrine of fair use, or fair dealing in certain jurisdictions, within
the context of AI-generated works. It investigates the factors that should be
considered in assessing whether a particular use can be deemed fair use when
AI-generated content is involved.
Artificial Intelligence And Copyright:
The main issue with the work generated by AI program is that whether the work
created by AI can be given Ownership under the Copyright law. The object of
determining this question is to ascertain right and liability of the work
protected under the copyright law. To answer this question, we have to analyse
the requirements for a work to get protection under the copy right law.
To
receive protection under copyright law, a creative work must meet several key
requirements. First and foremost, the work must be original, reflecting the
author's independent creativity and not be a direct copy of another work.
It
should also be fixed in a tangible medium of expression, which means it must be
recorded or stored in a way that can be perceived, reproduced, or communicated,
such as written documents, digital files, or audio recordings. While the work
should be original, it doesn't need to be exceptionally creative; a minimal
level of creativity is sufficient.
Copyright protection in India is governed by the Copyright Act, 1957[1]. To
receive protection under this Act, a creative work must adhere to certain
fundamental requirements. Firstly, it should be original, reflecting independent
creative effort and not be a mere copy of another work, as outlined in Section
13 of the Act.
The Act does not demand exceptional creativity, but it does
necessitate a basic degree of originality. Additionally, the work must be fixed
in a tangible medium of expression, making it perceptible, reproducible, or
communicable, as specified in Section 13(1)(a) of the Act.
This fixation
requirement is essential for copyright protection, and it's further emphasized
in Section 2(y). Copyright is automatically granted upon the creation and
fixation of the work, as stated in Section 17 of the Act, and formal
registration is not mandatory, though it can be beneficial for legal purposes.
Now the key point of consideration is whether the AI generated work original.
The court in Eastern Book Company v. D.B. Modak [2] held that the Copyright Act
1957 provides that for grant of copyright protection, the work must be original
i.e. it should originate from the author. In India, a work must involve a
minimum degree of creativity and not be a product of only skill and labour. The
Generative AI typically does not generate completely original data in the sense
of creating entirely new concepts or information that have never existed before.
Instead, it generates data that is based on the patterns and structures it has
learned from the training dataset. In other words, generative AI produces data
that is similar to what it has seen during training but is not necessarily
entirely novel or uniquely creative. The quality of the generated data depends
on the model's training and the diversity of the training dataset.
In essence,
generative AI is a tool for synthesizing data that fits within the boundaries of
what it has learned from existing data but does not possess true creativity or
the ability to invent entirely original content. This sums up one aspect of the
discussion that AI tools doses not possess originality in its work as Generative
AI tools train on data scrapped from millions of pre-existing sources and give
outputs based on a combination of these sources and their models.
Now the other aspect of creativity is to be taken into consideration while
ascertaining whether the AI tools fulfil the aspect of creativity. The
Generative AI can exhibit aspects of creativity within the constraints of its
training data. It can produce data that appears creative because it learns
patterns, styles, and structures from the training dataset, which it can then
recombine and generate in new and interesting ways.
However, this creativity is
often a form of "recombination" rather than true creative innovation or original
thought. Generative AI models can create novel variations, artistic
compositions, or even generate text that appears imaginative, but these are
ultimately based on the data they have been exposed to during training.
They
lack the capacity for true creative thinking, imagination, or the ability to
generate entirely original concepts or ideas that go beyond their training data.
In summary, generative AI can mimic creativity and produce novel outputs, but it
doesn't possess creative consciousness or independent, innovative thinking. It
generates based on patterns and knowledge extracted from its training data.
In 2021, an application was filed to the US Copyright office, for copyright
registration of a comic book consisting of text and images (created partly by a
human and partly by AI tool
Midjourney). The US Copyright Office refused to
grant copyright protection to that portion of the comic book which was created
by the AI tool. It also clarified that it will not register works produced by a
machine or mere mechanical process that operates randomly or automatically
without any creative input or intervention from a human author.[3]
The position in India may differ from the US Copyright office's view as the
copyright act gives protection to two kinds of works: primary works and
derivative works. Under the Indian Copyright Act, specifically the Copyright
Act, 1957, the concepts of primary work and derivative work are clearly
outlined. The term "original work" is used to describe what we've referred to as
a primary work, and it is defined in Section 13 of the Act. While the Act does
not explicitly define "original work," it suggests that copyright protection
applies to original literary, dramatic, musical, and artistic creations,
emphasizing that these works should be the result of the author's skill,
judgment, and labour.
In the context of derivative works, the Indian Copyright Act addresses them as
adaptations or modifications of the original work. In Section 2(ffa) of the Act,
a derivative work is referred to as an "adaptation" and is defined as a work
that has been altered, arranged, or transformed from the original work.
Furthermore, Section 14 of the Act specifies the exclusive rights of the
copyright holder, which includes the right to create adaptations or derivative
works based on the original work. This give us a case to argue that the output
produced by AI tools are derivative works and not copied from the primary work.
The another aspect of conferring copyright protection under the copyright law is
to determine who should be the author for the work to get protection under the
copyright law. Here our primary focus is on the authorship of the
computer-generated work as we are concerned with the ownership of the work
generated with the work generated by the AI.
The copyright act recognises authorship of computer-generated work. It
stipulates that the author of literary, dramatic, musical or artistic work which
are computer-generated is the "person who causes the work to be created."[4]
This provision was inserted into the Copyright Act through the 1994 amendments.
A reading of the Joint Committee Report on the Copyright ( Second Amendment
Bill), 1992 (Report) suggests that the provision was inserted keeping in mind
the rapid developments in technology, including artificial intelligence.
The
Committee acknowledged that through artificial intelligence, computers were
capable of giving new ideas, apart from what is fed into the system. The
Committee also discussed the need for protecting works generated by a computer
such as computerised music, animated films etc.
The Committee relied on the
Copyright, Designs and Patents Act, 1988 of the UK which grants protection to
"computer-generated works" i.e. work generated by a computer where there is no
human author. The Committee also acknowledged the need for distinguishing
between computer generated works and computer assisted works (where human
contributions can be easily identified).
Section 2(d)(vi), the Copyright Act 1957 provision that the person who causes
the computer-generated work to be created will be the author of the work. The
copyright law has interpreted natural person as author of the copyright work.
Further this can be better understood by the interpretation by the court in
respect to copyright law.
In
Camlin Pvt. Ltd. v. National Pencil Industries,[5] the Delhi High Court
elaborated the meaning of the term "author". The Courts stated that
"mechanically reproduced printed carton" was not a subject matter of copyright
for the reason that it was not possible to determine who the author of such
carton was. The Court further stated that "copyright is conferred only upon
authors or those who are natural person from whom the work has originated.
In
the circumstances the plaintiff cannot claim any copyright in any carton that
has been mechanically reproduced by a printing process as the work cannot be
said to have originated from the author. A machine cannot be an author of an
artistic work, nor can it have a copyright therein"
The Delhi high court in the case of
Navigators Logistics Ltd. v. Kashif Qureshi
& Ors.[6] rejected a copyright claim over a list compiled by a computer, on the
grounds of, inter alia, lack of human intervention.
In a case filed in late 2022,
Andersen v. Stability AI ltd.[7], three artists
formed a class to sue multiple generative AI platforms on the basis of the AI
using their original works without license to train their AI in their styles,
allowing users to generate works that may be insufficiently transformative from
their existing, protected works, and, as a result, would be unauthorized
derivative works. If a court finds that the AI's works are unauthorized and
derivative, substantial infringement penalties can apply.
Further, in 2020, the Copyright Office had recognized an AI tool, Raghav, as an
author of an art work produced by the AI tool, along with the developer of the
AI tool. This was seen as the first time that an AI tool was being recognized as
an author of a copyrighted work in India. However, subsequently the Copyright
Office issued a withdrawal notice, stating that the onus was on the applicant to
inform the Copyright Office about the legal status of the AI tool.[8]
This can be further understood with reference to an example i.e. when a work is
created by using a Microsoft program the person who puts his inputs in the
program is considered as the author of the program. It is not the program which
is the considered as the author, as the inputs are provided by the user of the
program and it includes their creativity and knowledge, while in the case of
work generated by AI the user does not puts in their knowledge, skill or
creativity it is the software which does all the task and generates the due
output.[9]
Therefore from the above we can draw a conclusion that in case of AI generated
work is given the authorship it would be against the legislative intend of the
legislature and also against the interpretation of the court.
From the above discussion now let us try to draw a conclusion that whether AI
generated work can be given ownership under the copyright law. As discussed
above to answer this question we have to analyse the requirement for a work to
get protection under the copyright law i.e. is the AI generated work original,
creative and can the authorship of the work conferred on anyone other then a
natural person.
The above discussion answers all the question helps us to draw
to a conclusion that the AI generated work does not fulfil any of the
requirement as provided under the copyright law. Therefore, the answer to the
question whether AI generated work can be given ownership under the copyright
law is in negative.
Faith Of Copyright Owner In The Presence Of AI: Determination Of Liabilities:
The question of the authorship to the AI software will further raise questions
as to the determination of the liabilities in the cases of infringement of the
work having copyright protection under the copyright act. Certainly. Under the
Copyright Act, there are several liabilities and offenses defined to protect the
rights of copyright holders and deter copyright infringement which protects the
rights and liability of the copyright owners.
The Indian copyright law imposes various liabilities under the copyright law
which are as following:
Civil liabilities are primarily governed by Section 51 of the Copyright Act.
This section addresses copyright infringement, making it clear that anyone who
reproduces, distributes, performs, displays, or adapts a copyrighted work
without the authorization of the copyright holder can be held civilly liable.
The consequences of such infringement may include civil penalties, such as the
payment of damages to the copyright owner, the issuance of injunctions to stop
further infringement, and the seizure or destruction of infringing copies.
Criminal liabilities, on the other hand, are addressed in Section 63 of the
Copyright Act. Engaging in activities like the production and distribution of
pirated or counterfeit copyrighted materials can lead to criminal liability,
including fines and imprisonment. This is a more severe form of liability that
may apply when infringement is carried out on a large scale or with intent to
profit significantly.
Furthermore, the Act also addresses circumvention of technological protection
measures, as outlined in Section 65A. This section imposes penalties for
intentionally circumventing technological measures put in place to prevent
unauthorized access to copyrighted works. Such measures are often used to
protect digital content and restrict unauthorized copying or distribution. Moral
rights, recognized in Section 57, grant authors and creators the right to object
to derogatory treatment of their works and claim authorship.
Violating these
moral rights can result in legal action and liability, especially when there is
unauthorized distortion or misuse of the original work. Online copyright
infringement, a growing concern in the digital age, is also addressed within the
Copyright Act, particularly in Section 52. This section holds internet service
providers and intermediaries accountable for hosting or providing access to
infringing content.
It may also specify safe harbour provisions for some service
providers, outlining their responsibilities and limitations in addressing online
infringement effectively. It's important to note that the specific liabilities
and penalties may vary from one country to another, as copyright laws are often
national in scope. Additionally, the Copyright Act may be periodically updated
to address new challenges and technologies related to copyright and intellectual
property.
Therefore, it's advisable to consult the most recent version of the
Copyright Act in a specific jurisdiction for precise details regarding
liabilities and offenses. These all the liabilities are vested on the author of
the work if the copyright protection is given to the AI generated work, then
another question arise that upon whom these liability will be vested.
The copyright protection serves as an incentive for the author to produce more
creative works using his skills, labour and judgement. If the AI is recognised
as an author and the AI generated works are protected under the copyright law,
then it would mean that "human creativity" and "machine creativity" are on the
same pedestal.
On the other hand, if AI-generated works are not protected by
copyright law, then it would necessarily mean that human creativity is preferred
over machine creativity. Preferring machine creativity over human creativity or
putting both at the same pedestal is likely to kill human creativity in the long
run.[10]
The question whether the AI generated work can be granted ownership under the
copyright law has already been discussed and the answer to which is in negative.
The another contention is that if the AI generated work infringes the right of
the copyright holder who would be liable for the infringement. In case where the
work does not have a defined owner then it is a difficult task to determine the
liability for the infringement of the work which has been already granted
protection under the copyright law.
If AI-Generated work is granted copyright then another issue will arise as to
the duration of the term for which the AI generated work would be granted
protection the term may be counted from the date of publication for a period of
50 or 60 years depending upon the laws of the countries. Under the copyright act
the protection is granted to the author for life time plus 60 years. AI not
being an human being if the protection is granted the determination of the
period for which the protection is to be continued would become a point of
dispute.
AI And The Doctrine Of Fair Use:
The doctrine of fair use is a fundamental legal principle in copyright law that
allows for the limited use of copyrighted material without the need for
permission or payment to the copyright holder under specific circumstances. It
is designed to strike a balance between protecting the exclusive rights of
copyright owners and fostering essential elements of a free and creative
society, including freedom of expression, education, critique, and knowledge
dissemination.
Several factors guide the determination of fair use, such as the
purpose of use, the nature of the copyrighted work, the quantity of the material
used, and the potential impact on the market for the original work.
If a use is
for purposes like criticism, commentary, news reporting, teaching, scholarship,
research, or parody, it is more likely to qualify as fair use. While fair use is
a critical element of copyright law, it is a complex, case-specific doctrine,
and courts examine each situation individually to determine whether a particular
use meets the fair use criteria. This doctrine is pivotal in promoting
creativity and free expression while still respecting the rights of copyright
holders.
The Indian Copyright Act, 1957, does not explicitly mention the "fair use"
doctrine as a distinct legal concept, as the U.S. copyright law does. Instead,
the Act uses the term "fair dealing," and it outlines specific exceptions and
limitations to copyright infringement in various sections.
These exceptions
allow for certain uses of copyrighted material without the copyright owner's
permission under specific conditions. For instance, Section 52 of the Indian
Copyright Act includes several exceptions that are akin to the fair use concept.
These include provisions for fair dealing for purposes such as research, private
study, criticism, review, news reporting, and education. These exceptions permit
the use of copyrighted material without authorization, provided that the use
falls within the limits prescribed by the Act and is consistent with fair
dealing principles.
In deciding fair use cases, courts must consider the following factors having
equal weight:
- The purpose and character of the use, including whether it is commercial, transformative, and non expressive;
- The nature of the copyrighted work;
- The amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
- The effect of the use upon the potential market for or value of the copyrighted work.[11]
The court while determining whether a work falls within the doctrine of Fair use
takes into consideration the above following factors, however while determining
whether a work falls within the domain of fair use the court considers each case
on individually as determining them on a set president would be ineffective.
Now let us discuss whether the work generated by the AI program falls within the
doctrine of fair use. For a work to get protection under the Doctrine of Fair
use it has to full fill the above-mentioned requirements. Let's assume that the
underlying text was copied in some way during the ChatGPT training process.
Let's further assume that outputs from ChatGPT are, at least sometimes,
derivative works under copyright law. If that is the case, do copyright owners
of the original works have a copyright infringement claim against OpenAI?
Not if
the copying and the output generation are covered by the doctrine of "fair use."
If a use qualifies as fair use, then actions that would otherwise be prohibited
would not be deemed an infringement of copyright.[12]
Therefore, while determining the work generated by AI program the court has to
test the output based on the above-mentioned conditions. Like if the AI program
is involved in commercial activities or transformative work the it would not
fall within the doctrine of fair use.
CONCLUSION:
In the rapidly evolving intersection of artificial intelligence and copyright
law, a multitude of intricate legal questions and challenges emerge. At the core
of this intersection lies the fundamental query of whether AI-generated works
can be granted ownership and protection under copyright law.
To address this
complex issue, one must examine the essence of originality and creativity within
AI-generated content, contemplate the question of authorship for AI creations,
and untangle the intricate web of determining liabilities in cases of copyright
infringement involving AI. AI-generated works often fail to meet the required
level of originality as mandated by copyright law, primarily deriving from
patterns and data-driven processes.
The absence of human consciousness and
subjective creative intent in AI-generated content makes it challenging to
designate authorship, raising questions about who should be recognized as the
author of these works and how authorship status influences copyright ownership.
Furthermore, when copyright infringement cases involve AI-generated works,
determining liabilities becomes a complex issue, with questions arising about
who should bear the responsibility for copyright violations.
In cases where
AI-generated works mimic creativity within the constraints of their training
data, they may exhibit recombination of existing patterns and structures, rather
than true creative innovation or original thought. If copyright protection is
extended to AI-generated works, it opens up discussions about the duration of
such protection, as AI lacks human attributes.
Additionally, the doctrine of
fair use, or fair dealing in certain jurisdictions, becomes a relevant
consideration when assessing the permissibility of using AI-generated content,
taking into account factors like the purpose of use, the nature of the work, the
quantity used, and the potential market impact.
Navigating the intricate confluence of artificial intelligence and copyright law
presents a myriad of challenges and questions. To address these complex issues
effectively, several key suggestions emerge. Firstly, there's a need for legal
clarity regarding the eligibility of AI-generated works for copyright
protection. This clarity should encompass defining the requisite level of
originality and establishing guidelines for authorship.
Specifically, AI itself
should not be considered an author, and authorship should be attributed to those
who develop, train, or deploy AI tools. Crafting a dedicated legal framework for
AI-generated works, along with international collaboration and standardization,
is essential to ensure consistent practices on a global scale.
Furthermore, fair
use guidelines must evolve to accommodate AI's role in content creation,
considering factors like purpose, nature of the AI-generated work, and market
impact. Encouraging responsible AI development is vital, with a focus on ethical
principles that prevent copyright violations and promote transparency.
Liability
and enforcement mechanisms should be reviewed and updated, making AI platforms
share the responsibility for monitoring and preventing copyright infringements.
Public awareness and education initiatives are necessary to ensure that
creators, users, and AI developers understand the legal nuances and ethical
considerations.
Continued monitoring and adaptation of copyright laws are
essential to keep pace with the rapid evolution of AI. Finally, efficient
dispute resolution mechanisms, such as specialized AI courts, can be established
to address copyright disputes involving AI-generated works fairly and promptly.
These suggestions collectively work to strike a balance between protecting
creative works and recognizing AI's distinctive role in content creation,
ensuring a fair and just legal framework for all stakeholders.
In conclusion, the intricacies of aligning existing copyright law with the realm
of AI-generated creativity reveal a delicate balance between protecting human
creators' rights and recognizing the unique nature of AI-generated works.
Currently, AI-generated works often do not fulfil the requirements for copyright
protection, leading to the conclusion that granting ownership under copyright
law is not feasible. The future of copyright law in the age of AI continues to
evolve, and addressing these complex questions will be crucial in shaping the
legal landscape.
The determination of liabilities in the presence of
AI-generated works, should they receive copyright protection, poses additional
challenges, as does determining the duration of copyright protection. The
application of the doctrine of fair use to AI-generated content will also depend
on specific circumstances and legal interpretations.
End-Notes:
- https://copyright.gov.in/documents/copyrightrules1957.pdf
- 2002 PTC 641
- https://www.copyright.gov/docs/zarya-of-the-dawn.pdf
- Section 2(d)(vi), the Copyright Act 1957
- AIR 1986 Delhi 444
- 254 (2018) DLT 307
- 3:23-cv-00201, (N.D. Cal.)
- Exclusive: Indian Copyright Office issues withdrawal notice to AI co-author | Managing Intellectual Property
- Andres Guadamuz, Artificial Intelligence and copyright - https://www.wipo.int/wipo_magazine/en/2017/05/article_0003.html
- V. K. Ahuja, Artificial Intelligence And Copyright: Issues And
Challenges - https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3864922
- Diana Bikbaeva, AI Trained on Copyrighted Works: When Is It Fair Use?
- Diana Bikbaeva, AI Trained on Copyrighted Works: When Is It Fair Use? -
https://www.lexology.com/library/detail.aspx?g=9da9d7a7-4e69-4342-99fd-2ac51b369569
Please Drop Your Comments