ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1768
AI-GENERATED CONTENT AND COPYRIGHT LAW: A COMPARATIVE ANALYSIS
OF THE U.S.A. AND UZBEKISTAN
Abdusattorov Shokhjakhon Jurabek ugli
Penn state law master’s degree, Legal assistant, Legal point law firm, Pennsylvania, USA
https://doi.org/10.5281/zenodo.15572587
Abstract. This article examines the evolving challenges that artificial intelligence (AI)
poses to traditional copyright frameworks, focusing specifically on AI-generated content. It
begins by defining AI-generated works and exploring why these outputs disrupt long-standing
legal principles, such as human authorship, originality, and creative expression. The article then
conducts a comparative legal analysis between the United States and Uzbekistan—two
jurisdictions with markedly different levels of technological development and legal
infrastructure.
In the U.S., copyright protection is explicitly reserved for human creators, as reinforced
by recent case law and the U.S. Copyright Office’s guidance. Conversely, Uzbekistan's legal
framework does not yet address AI-generated content, offering both challenges due to legal
uncertainty and opportunities for progressive reform. The article highlights the lack of legal
precedent in Uzbekistan and underscores the importance of establishing a coherent IP policy
that balances innovation and legal protection.
Through this comparative approach, the article aims to inform policymakers, scholars,
and practitioners on how different jurisdictions can adapt or reform copyright law in response to
AI advancements. It concludes by advocating for flexible, forward-looking legal strategies that
accommodate technological change while upholding the fundamental principles of intellectual
property.
Keywords: Artificial Intelligence (AI), AI-generated content, copyright law, intellectual
property, authorship, originality, United States law, Uzbekistan law, comparative legal analysis,
machine-generated works, legal reform, copyright protection, U.S. Copyright Office, Thaler v.
Perlmutter, fair use, digital creativity, AI and law, copyright challenges, emerging technologies,
legal frameworks.
Introduction: Defining AI-Generated Content and Its Challenge to Traditional
Copyright Frameworks
The advent of artificial intelligence (AI) has transformed numerous industries, from
healthcare and finance to transportation and education. Among the most intriguing—and legally
complex—applications of AI is its capacity to generate creative content. Text, music, images,
and even software code can now be autonomously produced by sophisticated AI systems such as
ChatGPT, DALL·E, Midjourney, and MusicLM. These works often resemble human-created
content in quality and creativity, raising urgent questions within the realm of intellectual
property (IP) law—particularly copyright.
AI-generated content
refers to works created, in whole or in part, by artificial
intelligence without direct human authorship. This may include an AI writing a poem, generating
a digital artwork, composing music, or producing computer code.
ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1769
The degree of human involvement may vary: some outputs are guided by human prompts
and edited afterward, while others are autonomously created by machines trained on massive
datasets.
1
Traditionally, copyright law has operated on a foundational assumption:
authorship is a
human endeavor.
It is a legal doctrine based on the idea that creative works are the result of
human intellect and personal expression. This assumption is enshrined in most international
treaties and national laws governing copyright, including the Berne Convention, which defines
an “author” implicitly as a natural person. AI-generated works fundamentally challenge this
principle by introducing content created by machines that lack consciousness, creativity in the
human sense, or legal personhood.
Why AI-Generated Content Challenges Traditional Copyright
1.
Absence of Human Authorship
The primary challenge posed by AI-generated works is the
lack of a human author
—a
core requirement for copyright protection in most jurisdictions. For example, in the United
States, the Copyright Act does not define "author," but case law and regulatory guidance have
consistently interpreted it to mean a human being. The U.S. Copyright Office has explicitly
rejected registrations for works created solely by AI, stating that copyright only protects “the
fruits of intellectual labor that are founded in the creative powers of the [human] mind.” This
principle was reaffirmed in
Thaler v. Perlmutter
, where the federal court upheld the U.S.
Copyright Office’s denial of copyright to a piece of art generated entirely by an AI system
without human input.
2
AI challenges this requirement by creating works that are often indistinguishable from
human-made creations, yet lack a definable human author. If no human can be identified as the
creator, then the work may fall outside the scope of copyright protection, leaving it in the public
domain or legally unprotected.
2.
Questions of Ownership and Liability
In situations where AI is used collaboratively or as a tool—such as a designer using
Midjourney to create illustrations or an author using ChatGPT to co-write a story—the question
becomes:
who owns the output?
Is it the programmer of the AI? The user who gave the
prompt? The company that developed the model?
Traditional copyright law is ill-equipped to resolve these questions. Unlike works for
hire, AI lacks intent or legal standing, so assigning ownership becomes legally uncertain. This
opens the door to disputes over rights, revenue, licensing, and infringement liability, especially
when AI-generated content is commercialized.
3.
Lack of Originality and Human Expression
Another foundational requirement for copyright is
originality
, which implies some
minimal degree of creativity. In most jurisdictions, originality must stem from a human author’s
independent expression. Courts have historically rejected copyright protection for works created
1
U.S. Copyright Office,
Copyright Registration Guidance: Works Containing Material Generated by Artificial
Intelligence
(Mar. 2023), available at
2
Thaler v. Perlmutter
, No. 1:22-cv-01564 (D.D.C. Aug. 18, 2023).
ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1770
without sufficient human creative input, such as the infamous
monkey selfie
case (
Naruto v.
Slater
), where the court ruled that animals cannot hold copyrights.
3
AI-generated content may produce impressive results, but those results are ultimately the
product of algorithms and training data—not human expression. Even when a user inputs a
prompt into an AI model, the argument remains whether that input constitutes sufficient creative
control to warrant authorship. If originality and authorship must stem from a human source, then
works created autonomously by AI may not qualify for protection under traditional copyright
systems.
4.
Training Data and Fair Use Concerns
Beyond the authorship of outputs, there are also significant
copyright issues with AI’s
inputs
. AI models like ChatGPT and Stable Diffusion are trained on enormous datasets, many of
which contain copyrighted materials such as books, images, or music scraped from the internet.
Content creators have raised concerns that their copyrighted works were used without consent or
compensation, potentially violating copyright law.
The key legal question here is whether using such works in training datasets constitutes
fair use
or infringement. In the U.S., courts have not yet issued definitive rulings, but multiple
lawsuits are pending. In other jurisdictions, including Uzbekistan, the legal framework to address
this issue is still underdeveloped or silent.
Moral Rights and AI
A more philosophical, yet still legally relevant, challenge concerns
moral rights
—
particularly the right of attribution and integrity, which are recognized in many legal systems. If
an AI generates a work, can anyone be attributed as its author? Should the person who trained or
prompted the AI receive recognition, or is that fundamentally misaligned with the idea of
personal expression?
Moreover, when a user modifies AI-generated content, does it constitute a violation of
any moral rights? These are new legal questions that traditional frameworks struggle to address,
especially as AI content is widely shared, remixed, and repurposed across the internet.
4
Why Comparative Analysis is Essential
As AI becomes more pervasive globally, it is crucial to understand how different
jurisdictions are responding to these challenges. The United States, with its advanced tech sector
and robust legal precedents, offers an active but cautious approach—rejecting copyright for AI
works, yet dealing with emerging cases in real time. On the other hand, Uzbekistan is at an
earlier stage in both AI development and IP law reform, creating a unique opportunity to craft
forward-looking legal frameworks that address AI authorship from the outset.
By comparing these two legal systems, this article seeks to explore both the gaps and
potential paths forward in adapting copyright law to the realities of AI-generated content.
As artificial intelligence (AI) becomes increasingly capable of generating creative
outputs, copyright law is being tested in unprecedented ways. While the challenges AI poses to
traditional copyright frameworks are global, national legal systems vary significantly in how
3
Naruto v. Slater
, 888 F.3d 418 (9th Cir. 2018).
4
U.S. Copyright Office,
Zarya of the Dawn Registration Decision Letter
(Feb. 21, 2023), denying copyright to
images generated by Midjourney due to lack of human authorship.
ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1771
they respond to these challenges. This article seeks to conduct a comparative legal analysis of
two distinct jurisdictions: the
United States
, a highly developed and precedent-driven legal
system with a proactive intellectual property (IP) framework, and
Uzbekistan
, an emerging legal
jurisdiction undergoing digital and legal transformation.
The purpose of this comparative study is to examine how each country is currently
addressing—or failing to address—the issue of
AI-generated works and their copyright
status
. By comparing the legislative language, case law, regulatory guidance, and legal
interpretations in the U.S. and Uzbekistan, the study aims to:
•
Identify key similarities and differences in legal treatment of AI-generated content,
•
Evaluate how each system defines authorship and originality in the context of AI,
•
Assess the degree of preparedness and adaptability in each jurisdiction’s copyright law,
and
•
Propose recommendations for policymakers, especially in developing countries like
Uzbekistan, on how to create more AI-inclusive IP frameworks.
The U.S. Copyright Act does not explicitly mention AI, but the term "author" has been
consistently interpreted to refer to
natural persons
. The U.S. Copyright Office reaffirmed this
interpretation in its 2023 guidance, stating that “only works created by human beings are eligible
for copyright.” It further clarified that content generated by AI without substantial human
creative input does not qualify for protection.
The landmark case of
Thaler v. Perlmutter
(2023) reinforced the human authorship
requirement. Stephen Thaler, an AI researcher, attempted to register a piece of visual art created
by his AI system “Creativity Machine.” The court upheld the Copyright Office’s denial, ruling
that authorship must originate from a human mind.
Additionally,
Naruto v. Slater
, though not directly involving AI, is often cited as
precedent.
The case concluded that animals cannot own copyrights, reinforcing the principle that
non-human entities (whether a monkey or an AI) cannot be authors.
Beyond outputs, there is also growing legal debate in the U.S. over how copyrighted
materials are used to
train AI models
. Companies like OpenAI, Stability AI, and GitHub have
been sued for allegedly using copyrighted texts, code, and images in their training data without
permission. Whether this constitutes “fair use” remains an open question in U.S. courts, and the
outcome will significantly shape the future of AI training practices.
5
The U.S. approach reflects a conservative interpretation of copyright law, grounded in
long-standing principles. While it may limit the ability of AI creators to claim copyright over
machine-generated content, it also avoids the legal and ethical complexity of granting legal
rights to non-human entities. However, this also risks stifling innovation, as creators and
developers may hesitate to invest in AI content if it cannot be legally protected.
Uzbekistan: A Legal System in Transition
5
World Intellectual Property Organization (WIPO),
Revised Issues Paper on Intellectual Property Policy and
Artificial Intelligence
(May 2020), available at
ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1772
Unlike the U.S.,
Uzbekistan does not yet have a clear or explicit position
on AI-
generated content within its copyright law. However, it presents a unique case study of a legal
system that is still evolving—offering both opportunities and challenges.
Current Legal Framework
Uzbekistan’s copyright regime is primarily governed by the
Law on Copyright and Related Rights
, adopted in 1996 and amended in subsequent years.
Like many post-Soviet systems, the law emphasizes authorship, originality, and moral rights, but
it does not contain language that anticipates the complexities introduced by AI.
The law defines an “author” as a
natural person
who created a work of science,
literature, or art, similar to the U.S. framework. However, the absence of case law or regulatory
guidance on AI-generated content leaves the issue open to interpretation.
Institutional Capacity and Modernization Efforts
The government of Uzbekistan has recognized the importance of digital innovation and
intellectual property. Initiatives such as
“Digital Uzbekistan 2030”
aim to modernize the legal
and technological landscape.
Moreover, the
Intellectual Property Agency under the Ministry of Justice
has begun
updating IP procedures and raising awareness about the impact of AI on creative sectors.
6
However, these efforts remain at an early stage. There are no known court cases or
official legal opinions in Uzbekistan specifically addressing the copyright status of AI-generated
works or the legality of using copyrighted content in training datasets.
Potential Risks and Opportunities
The lack of explicit regulation creates legal
uncertainty, which can discourage investment and innovation in AI development. At the same
time, this legal vacuum offers
a valuable opportunity
for Uzbek lawmakers to proactively
shape future legislation by studying global trends and avoiding the pitfalls seen in other
jurisdictions.
For instance, Uzbekistan could adopt a flexible model where copyright protection is
granted to the
human user who meaningfully contributes
to the creation of AI-generated
content, thereby incentivizing creativity without compromising core legal principles.
Comparative Observations
Issue
United States
Uzbekistan
Definition of Author
Human only
Human only (statutory)
Copyright for AI-Generated
Works
Not eligible without human
input
No explicit guidance yet
Legal Precedents
Extensive (e.g., Thaler,
Naruto)
None currently
AI Training & Fair Use
Active litigation
No cases or official stance
Readiness for AI & IP Law
Reform
High (but conservative)
Low to moderate (but open to
change)
6
Law of the Republic of Uzbekistan “On Copyright and Related Rights,” No. 257-I of July 20, 1996 (as amended),
available (in Uzbek) at
ISSN:
2181-3906
2025
International scientific journal
«MODERN
SCIENCE
АND RESEARCH»
VOLUME 4 / ISSUE 5 / UIF:8.2 / MODERNSCIENCE.UZ
1773
Conclusion
The comparison between the United States and Uzbekistan reveals both
divergence and opportunity in how legal systems are responding to the rise of AI-generated
content. The U.S. has taken a firm stance that excludes non-human authors from copyright
protection, relying on historical doctrine and evolving case law to address emerging disputes.
While this approach offers legal clarity, it may also hinder future innovation, especially as AI
becomes more integrated into creative industries.
Uzbekistan, by contrast, is in a formative stage. Its copyright laws do not yet address AI,
but the nation’s commitment to digital transformation and legal modernization positions it to
adopt more adaptive and forward-thinking policies. This legal flexibility could allow Uzbekistan
to become a model for AI-inclusive copyright regulation in developing countries, particularly by
striking a balance between protecting human creativity and encouraging technological progress.
As AI continues to blur the lines between human and machine creativity, it is imperative
for countries to revisit traditional legal frameworks and consider whether, how, and to what
extent AI-generated works should be protected under copyright law. Through comparative legal
analysis, policymakers and scholars can better understand the global landscape and develop
coherent strategies that align legal protection with technological realities.
REFERENCES
1.
U.S. Copyright Office. (2023).
Copyright Registration Guidance: Works Containing
Material
Generated
by
Artificial
Intelligence
.
Retrieved
from
https://www.copyright.gov/ai
2.
U.S. Copyright Act, 17 U.S.C. § 102(a) (2022).
3.
Thaler v. Perlmutter
, No. 1:22-cv-01564 (D.D.C. 2023).
4.
Naruto v. Slater
, 888 F.3d 418 (9th Cir. 2018).
5.
Andersen v. Stability AI Ltd., No. 3:23-cv-00201 (N.D. Cal. 2023).
6.
U.S. Copyright Office. (2023).
Zarya of the Dawn
Registration Decision Letter.
Retrieved from https://www.copyright.gov/ai
7.
Law of the Republic of Uzbekistan “On Copyright and Related Rights,” No. 257-I, July
20, 1996 (as amended). Retrieved from https://lex.uz/docs/10857 (in Uzbek).
8.
Intellectual Property Agency under the Ministry of Justice of the Republic of Uzbekistan.
(n.d.).
Tasks and Strategic Goals
. Retrieved from
9.
Cabinet of Ministers of Uzbekistan. (n.d.).
Digital Uzbekistan 2030
Strategy. Retrieved
from
10.
World Intellectual Property Organization (WIPO). (2020).
Revised Issues Paper on
Intellectual
Property
Policy
and
Artificial
Intelligence
.
Retrieved
from
https://www.wipo.int/edocs/pubdocs/en/wipo_pub_450_2020.pdf
11.
Jütte, B. J. (2021). The Artificial Author? Copyright and Algorithm-Generated Works.
Journal
of
Intellectual
Property
Law
&
Practice
,
16(2),
95–103.
https://doi.org/10.1093/jiplp/jpaa196
