Artificial Intelligence And Copyright Law - Issues Of Authorship, Ownership, And Originality

Author: LawSphereX

Category: Intellectual Property

Published: 2026-08-23T18:11:54.77554+00:00

Updated: 2026-08-26T17:12:04.388+00:00

ARTIFICIAL INTELLIGENCE AND COPYRIGHT LAW:

ISSUES OF AUTHORSHIP, OWNERSHIP, AND

ORIGINALITY

Himanshi Khurana, Kalyan Law College, Bhilai (C.G)

1. ABSTRACT

In today's rapid advancement developing technological society artificial intelligence is developing exponentially on a large scale. Al has disrupted the fundamental principles of intellectual property law partially the doctrine of originality, authorship and ownership in which struggle to assist Al generated creativity. Tools such as ChatGPT, Google Gemini, Mid journey and cloud Al now autonomously produce literacy, artistic and musical works with minimal human interference, making current laws insufficient and creating a need to urgently review and inflate legal principles.

This paper examines the flow critical dimensions of the AI-copyright interface, first it studies the legal position of Al generative work and explains that not giving copyright protection to such work create major legal gap leads to misuse, negatively affect creative industries in Thaler vs Perlmutter, AI vs copyright office denied copyright to Al generated work. Similarly, India's copyright act to Al generated work on Al generated content creates a legal gap. Second, the paper critically examines the traditional human authorship doctrine and questions whether existing standards of originality and creativity remain adequate in limited human involvements to prompt engineering selecting ai generated output. Third, the paper examines ownership and licensing issues related to ai generated content by analyzing the competing rights of developers, user and ai platforms.

Also evaluates the work for giving doctrines or sui generis can effectively regulate ai generated works. Fourth, the paper examines ai, copyright infringement and liability arising from the use of copyrighted datasets in ai training, analyses the leading case The New York times vs Open Al and Microsoft and proposes a structured liabilities framework for developer's platforms and end users using a doctrinal and comparative approach the papers argues that India must amend its copyright act, 1957 to address AI related authorship, ownership and liability. Keywords: Artificial Intelligence and Copyright; Navigating Authorship; Ownership and Originality in the Age of Generative AI.

2. INTRODUCTION

2.1 Historical Context and Evolution.

Copyright law started in the 18th century with the Statute of Anne, 1710 in England. This law said that original literary and artistic work created by a person should get legal protection so that others cannot copy them without permission. The basic idea of copyright law was that a human being creates original work and therefore deserves rights over it. In the 20th century, copyright law expanded to protect books, music, painting, films, and computer software. However, one important rule always remained the same: copyright protection was mainly given to works created by humans. International agreements such as the Berne Convention, TRIPS Agreement, and WIPO Treaties also followed the same idea that creative work comes from human effort and creativity. Today, the development of Artificial Intelligence has created new challenges for copyright law.

Modern AI systems like Chat GPT, Google Gemini, Mid Journey, AI Claud can create text, images, music and video on their own. Because of this, an important question arises: if AI creates something, who is the real author? It is also unclear whether AI generated work should get copyright protection and who should own those rights – the developer, the user or no one. AI has also raised issues of copyright infringement and liability. Copyright infringement happens when someone uses or copies protected work without permission. AI systems are trained using large amounts of data from the internet, which may include copyright books, articles, artworks, and music. This creates a debate about whether using such material for AI training is legal or not. Another problem is that AI sometimes produces content very similar to existing copyrighted works.

For example AI may create an image similar to a famous artist’s style or text similar to a published book. In such cases, it becomes difficult to decide who is legally responsible. Since AI is not considered a legal person, the responsibility may fall on the developer, the company or the user using the AI system. Therefore, AI technology has challenged the traditional rules of copyright law, especially regarding human authorship, originality, ownership, infringement and legal liability.

2.2 Key Definition.

● Generative AI: AI system that creates content like text, images and music.

● Authorship: Legal recognition of the creator of a work. ● Infringement: Unauthorized use or copying of copyrighted work. ● Training Data: Data used to train AI models.

2.3 Existing Evidence and Current Legal Landscape.

Recent legal developments show that current copyright laws are not fully prepared to deal with AI-generated content. In the United States, the United States Copyright Office stated in its 2023 guidance that copyright protection requires “HUMAN AUTHORSHIP”. Because of this, purely AI-generated artworks cannot receive copyright registration. Several major legal cases, such as Getty Images vs Stability AI, questioned whether AI models use copyrighted content without permission or payment to creators. The UK Copyright Act also does not fully recognize copyright protection for AI- generated works that lack human involvement, However, some countries like China and the United Arab Emirate (UAE) have started giving limited recognition to AI- generated outputs. These different approaches in different countries show that there is no uniform global rule for AI and copyright law.

Therefore, there is a strong need for clear and harmonized legal frameworks to address AI- related copyright issues.

2.4 Statement of the Problem.

Although many academic studies have discussed the impact of AI on areas like music, visual art, and journalism, there is still limited research on the combined issues of AI authorship, ownership, its legal status and infringement liability. Most studies focus only on one country’s legal systems instead of comparing different international approaches. In addition, the practical impact of AI on stakeholders such as AI companies, creators, digital platforms, and consumers has not been fully examined, the rise of AI has blurred the boundary between human creativity and machine-generated expression. Therefore, this research aims to fill these gaps by analyzing legal principles, international perspective and real cases studies related to AI and copyright law.

2.5 Objectives and Scope.

Primary Objectives

● To study the legal status of AI–generated literary, artistic and musical works under copyright law. ● To examine the requirement of human authorship and its relevance to AI – generated content. ● To analyze ownership and licensing issues related to AI-created works. ● To evaluate copyright infringement and liability in AI-generated content. ● To suggest a legal framework that balances innovation, creator rights, and public interest. Scope of the Study This research focuses on international copyright law, especially in the US, UK, EU, India and emerging countries, covers literary, artistic, musical and cinematographic works protected under copyright law and examines different AI systems, from rulebased AI to advanced deep learning models

3. DOCTRINAL RESEARCH METHODOLOGY

3.1 Framework of the Study

This research adopts a mixed-method approach combining doctrinal legal analysis, comparative case studies, and expert insights to examine the relationship between AI and copyright law. The study mainly focuses on analyzing copyright statutes, legal regulation, judicial decisions and legislative development across different countries. In addition, landmark cases and important regulatory decisions have been comparatively examined to understand the evolving legal challenges of AI- generated content.

3.2 Material and sources

This research is based on both primary and secondary sources to understand the legal issues related to AI and copyright law.

Primary Sources Primary sources include official copyright laws, court judgement, government guidelines, and important legal cases. The study examined laws such as the US Copyright Act,1988, and the EU Copyright Directive. It also analyzed important court decisions and recent AIrelated cases like Getty Images vs Stability AI and The New York Times Co. vs Open AI. Government documents and policy guidelines related to AI and copyright were also reviewed. Secondary Sources Secondary sources include academic journals, research articles, legal books, commentaries, and industry reports. Report and publication from organizations such as the World Intellectual Property Organization, Electronic Frontier Foundation, and Creative Commons were used to understand different legal opinions and modern development in AI and copyright law.

3.3 Research Methodology and Analytical Procedure Examination of Legal Doctrines

The research examined copyright law and regulation from different countries. It focused on important legal concepts such as authorship, originality, ownership and infringement to understand how existing copyright laws apply to AI-generated content. The study also identified legal uncertainties related to AI authorship. Comparative Analysis of Landmark Cases Important copyright cases related to AI were comparatively analyzed. These cases included disputes regarding AI training data, royalty claims, and copyright registration refusals. Cases from different countries and creative sectors were selected to understand various legal approaches. Synthesis of Legal Principles The research combined legal principle and judicial interpretation to identify common patterns, contradictions, and gaps in copyright law related to AI- generated works.

It also examined how courts and lawmakers interpret the requirement of human authorship.

Analysis of Stakeholder Perspective The study reviewed opinions, policy papers, and reports from AI developers, creative industry professionals, consumer groups and policy organizations to understand different viewpoints and proposed solutions regarding AI and copyright law.

3.4 Research Tools and Analytical Instruments

This research used various legal research tools and analytical methods to collect and examine information related to AI and copyright law. ● Legal Research Database: Sources such as LexisNexis, Westlaw, Google Scholar, and the World Intellectual Property Organization database were used to access laws, legal cases, journals, and policy documents. ● Comparative Legal Frameworks: A comparative framework was prepared to examine differences in authorship rules, ownership rights, infringement standards, and legal remedies across different jurisdictions. ● Cases Citation and Interpretation Tools: Legal research tools were used to study important judgements, judicial reasoning, and court decisions related to AI and copyright disputes.

● Thematic Analysis Method: Information was systematically categorized to identify major legal themes, judicial interpretations, and policy arguments regarding AI-generated content.

3.5 Reliability and Validity of the Study

This research ensured reliability and validity by using authentic and verified legal sources, including official government websites, court records and recognized legal databases. The study examined copyright laws from different jurisdictions and legal systems to provide a broader and more balanced analysis of AI and copyright law. Academic opinions and legal commentaries were also compared with judicial decisions to identify consistent legal interpretations and expert consensus. In addition, the research considered the historical development of copyright law to understand how

legal principles have evolved with technological advancements, particularly in the field of Artificial Intelligence.

4. AI &COPYRIGHT: KEY FINDINGS.

The following section presents the key findings and legal analysis of AI- generated content under copyright law, focusing on authorship, ownership, infringement and related legal challenges.

4.1 Legal Status of AI- Generated Works

Finding 1: Uncertainty in Copyright Protection for AI- Generated Content. The research found that the legal status of AI-generated works remains uncertain across most jurisdictions. In the United States, the United States Copyright Office has clearly stated that copyright protection requires “HUMAN AUTHORSHIP”. An analysis of copyright registration refusals between 2022 and 2023 showed that works created entirely by AI were denied protection because they lacked human creative involvement. Similarly, the UK Copyright Act also limits copyright protection to works created by humans. In the European Union, courts have emphasized that an “author” must make free and creative choices while creating a work. These legal approaches show that current copyright law is mainly designed for human creators and is not fully prepared to address AI-generated content.

As a result, a major legal dilemma arises. Either AI systems must be recognized as legal authors, which is both legally and philosophically controversial, or AI-generated works may remain outside copyright protection altogether. This creates uncertainty using AI technologies, as there is still no clear legal framework defining ownership and protection of AI- generated content. Based on the finding and research, several legal and policy recommendations can be proposed to address the growing challenges of AI in copyright law. Existing copyright frameworks should be updated to clarify issues related to authorship, ownership,

infringement and liability. Government and policymakers should develop balanced legal frameworks that protect the rights of human creators while also encouraging technological innovation and AI development. A clear licensing mechanism and fair compensation system should be introduced for copyrighted materials used AI training datasets. The research also highlights the need for greater international cooperation to establish harmonized global standards for AI-generated content should be promoted to prevent misuse, plagiarism and digital exploitation. Finally, public awareness and digital literacy regarding AI-generated works and copyright rights should be strengthened to ensure responsible and lawful use of AI in the creative sector. Analysis of the Humans Authorship Requirement. Finding 2: Challenges to the Humans Authorship Requirement.

The research found that the requirement of human authorship faces several legal and interpretative challenges in the age of AI. First, cases such as Naruto vs Slater established that non-human creators cannot hold copyright protection, even when they create uncertainty for AI-generated works because AI systems also non-human creator. Second, copyright decisions related to photography show that even minimal human creativity, such as selecting the subject, angle or timing may qualify for copyright protection. This suggests that limited human involvement in directing AI systems could potentially satisfy the requirement of authorship. Third, the growing autonomy of AI systems challenges traditional theories of authorship that are based on human intention, creativity, control.

Judicial analysis shows that courts continue to struggle in clearly defining why only humans should receive authorship rights while completely excluding AI participation in the creative process. In light of growing use of AI in creative processes, copyright law should establish clear guidelines regarding the requirement of human authorship in AI-generated works. Legal frameworks should recognize meaningful human creativity, control and directions in the

use of AI systems while ensuring that fully autonomous AI outputs are appropriately regulated. Such clarity would help reduce legal uncertainty and create a balanced approach between technological innovation and copyright protection. Moreover, Education institution should promote responsible use of AI-generated materials, Policymakers should balance innovation with creator protection, Court should adopt flexible interpretation of authorship in AI-assisted creations, Human creativity should remain an important factor in copyright protection, Fair compensation mechanism should be introduced for creators whose works are used in AI training.

Therefore, suggestion explain how the existing problems can be solved and how the legal system can be improved in the future also demonstrate the practical importance of the research by proving recommendations for improving existing laws, policies and legal framework related to AI and copyright protection. Ownership and Licensing Mechanisms. Generative AI has transformed the creation of content such as art, music, literature and images. However, this technological advancement has also created a major legal question—Who actually owns AI- generated content? – The AI developer who created the system, user who gave prompt, copyright holder whose works trained the AI or should such content belong to the public domain? Currently, there is no universal legal consensus regarding ownership licensing policies, which creates uncertainty for creators, businesses and legal systems.

to understand this issue, doctrinal analysis revealed 4 major competing ownership theories. Finding 3: Four competing ownership allocation theories emerged from doctrine analysis; 1. Clear legal Framework for Ownership: Government should create a uniform legal framework defining who owns AI- generated content – developers, users or both. This will reduce confusion and future legal disputes. 2. Humans Contribution Test: Ownership should depend on the level of human creativity involved in promoting or editing AI outputs. “More human creative control = stronger ownership rights”.

3. Transparent Licensing Policies: AI companies should provide simple and transparent Terms of Services regarding: ● Commercial use, ● Copyright Ownership, ● Liability, ● Revenue Rights 4. Protection for Training Data Creators: Artists, writers, musicians and other copyright holders whose works are used for training should receive— ● Consent Mechanisms, ● Attribution, ● Compensation Models. 5. Creation of Hybrid Ownership Model: A balanced/shared ownership model between AI developers and users may be introduced for certain high-value AIgenerated works. 6. International Harmonization: Countries should work towards globally consistent AI copyright standards because AI platforms operate internationally. What will happen if these suggestions are implemented? ● Legal uncertainty in AI copyright will reduce, Fewer copyright lawsuits and Ownership disputes will arise.

● Creators and companies will better understand their rights. Innovation in AI technology can continue safely. Artists and copyright holders may receive fair protection and compensation. ● Public trust in AI-generated content will increase, Global AI markets will become more stable and commercially reliable.

Copyright Infringement and Liability The rapid growth of generative AI has created serious concerns regarding copyright infringement and legal liability. AI systems are trained on massive amounts of online data, including copyrighted images, books, music and artistic works. This has raised an important legal question: Can AI companies use copyrighted material for training without permission and who will be liable if AI- generated outputs infringe copyright? Finding 4: Training AI systems on copyrighted content generates 2 distinct infringement issues: First, the use of copyrighted material during AI training may constitute unauthorized reproduction of protected works, thereby infringing reproduction rights. AI developers often collect large datasets from the internet without obtaining licenses or consent from copyright holders.

Second AI- generated outputs that closely resemble existing copyrighted works may amount to derivate works, potentially infringing adaption rights. This issue becomes more complex when AI systems autonomously generate content similar to their training data. A major example is Getty Images vs Stability AI, where Getty Images alleged that

stability AI copied millions of copyrighted images without authorization for AI training purposes. The case highlights the growing conflict between technological innovation and copyright protection. The analysis further reveals uncertainty regarding legal liability. Courts must determine whether AI training qualifies as “fair use” under the doctrine of transformative use, whether AI companies should bear liability of infringing outputs and whether prompt users may face secondary liability for copyright violations caused through AI-generated content.

This key finding determines : AI companies should obtain proper licenses before using copyrighted works for AI training, Clear legal standards should be established regarding fair use in AI training activities, Copyright laws should defines the liability of AI developers, platforms providers and prompt users separately, AI systems should include transparency mechanisms to identify the sources of training data and generated outputs, Government should introduce balanced regulations that protect both innovation and the

rights of copyright holder.

JURISDICTION

AI AUTHORSHIP

TRAINING

FAIR USE STATUS

LIABILITY

UNITED STATES

Requires human Uncertain Being tested authorship UNITED KINGDOM No clear recognition Developer liable Not granted EUROPEAN UNION Must be human Under review Likely restricted

INDIA

No specific

AI

Legally uncertain Limited and case authorship law specific Comparative Jurisdiction Analysis.

TABLE ANALYSIS

The comparative analysis shows that most jurisdictions continue to prioritize human authorship in copyright law. While the US, UK, EU and India have begun developing clearer approaches towards AI regulation, India still lacks specific legislation governing AI- generated works. Liability regarding AI training and infringement remains legally uncertain in India and fair use protections are interpreted narrowly on a case-by-case basis. This legal uncertainty highlights the urgent need for a comprehensive AI copyright framework in India. ● Countries should introduce specific AI copyright legislation. ● AI companies should use licensed training datasets. ● International corporations are necessary for uniform AI regulation. ● Clear liability rules should be established for developers and users.

5. WHAT DO MY FINDINGS SHOWS?

This finding is located within two competing approaches to copyright law in the context

of Artificial Intelligence. The Protectionist Approach advocates strict applications of existing copyright principles. It maintains that AI systems cannot qualify as authors, that the use of copyrighted works for training without authorization constitutes infringement and that strong legal safeguards are necessary to protect human creators from AI-driven market displacement. While this approach preserves the doctrinal integrity of traditional copyright law, it raises concerns regarding the lack of legal protection for AI-generated outputs, which may discourage the development and deployment of AI technologies. On the other hand, the Innovation approach supports more flexible and adaptive interpretations of copyright law.

It argues that minimal human inputs may be sufficient to establish authorship, that training of AI models may fall within the scope of transformative fair use and that overly rigid enforcement may hinder technological progress and innovation. However, this approach also risks undermining the economic and moral interest of human creators by diluting the value of original works. Both approaches, in their extreme forms, fail to adequately balance the competing interests of innovation and creator protection. Accordingly, this research proposes a Tiered Protection Framework, which allocates copyright protection based on the level of human creative inputs involved.

Works involving substantial human direction or creative control may be granted full copyright protection, whereas autonomously generated outputs may receive limited protection, such as protection against direct commercial reproduction while still permitting adaptation and fair use. Furthermore, the use of openly licensed training data, when properly attributed, should be permitted without infringement liability. This calibrated framework seeks to balance incentivizing innovation with safeguarding the rights and interest of human creators in the evolving AI ecosystem. “Law should not be fully strict or fully free, instead protection should depend on how much human creativity is involved “. Practical Implications ● AI Developers: Legal uncertainty increases business risk, needs clear rules on ownership and liability.

● Content Creator: Limited protection against AI- assisted or AI- competitive works under current copyright law. ● Rights Holders: AI training liability is unclear; licensing is becoming necessary for protection. ● Consumer: Unclear copyright status AI outputs creates legal and commercial risk in usage.

6. CONCLUSION

This research finds that current copyright law is not adequately equipped to deal with Artificial Intelligence in creative works. The traditional requirement of human authorship creates rigid outcomes, leaving AI-generated content and related legal issues like infringement and licensing unclear. Comparative analysis shows inconsistent approaches across jurisdictions, leading to legal uncertainty and lack of international harmonization. The study concludes that copyright reforms are necessary. First, the human vs nonhuman authorship distinction should be replaced with a framework based on the degree of human creative inputs. Second, training data use must be clearly regulated through fair use clarification or statutory licensing to protect rights holders. Third, international coordination is essential to reduce legal fragmentations.

Overall, copyright law must evolve to balance innovation and creator protection. Recommended reforms include recognizing AI as a non-author, introducing statutory licensing for training data and using technological tools like watermarking and provenance tracking to ensure transparency and fairness in the AI ecosystem.

REFERENCES

● Berne Convention for the Protection of Literary and Artistic Works (1886, amended 1971). World Intellectual Property Organization. ● Cariou V. Prince, 714 F. 3d694 (2nd Cir 2013). ● United States Copyright Office. (2023). Guidance on Copyright Registration and Artificial Intelligence. Washington DC. ● Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019. ● Feist Publication, Inc. V. Rural Telephone Services Co. ,499 U. S. 340 (1991). ● Getty Images (US), Inc. V. Stability AI, Inc. , Case No. 23-20910 (N. D Cal. 2023). ● Intellectual Property Office, UK. (2022). Artificial Intelligence and Intellectual Property: Policy Position Statement. London. ● Naruto v. Slater, 888 F. 3d 418 (9th Cir. 2018). ● New York Times Co. V. Open AI, Inc. , Case No. 24-11557 (S. D. N. Y. 2024). ● Samuelson, P. (2021). “Generative AI Meets Copyright”.

Journal of the Copyright Society of the U. S. A. , 68(3), 209-250. ● The Copyright, Designs and Patents ACT 1988 (UK). ● Title 17 of the United States Code: Copyright Law (1976, amended 2023). ● World Intellectual Property Organization. (2023). Artificial Intellectual and Copyright: Challenges and Opportunities. Geneva.