AI and Copyright: A Modern Debate

Introduction

In recent years the monumental growth and speculation over artificial intelligence has sparked perhaps the most urgent debate of modern times, AI and Copyright. This debate highlights the most important question that we need to answer, the answer which will change the course of artificial intelligence and will determine how AI is going to be used by us and our future generations.

According to a UN Trade and Development (UNCTAD) report, the global AI market is projected to surge from $189 billion in 2023 to $4.8 trillion by 2033 , a staggering 25-fold increase within a decade. A major driver of this expansion is the rise of generative AI, the technology that powers content-creation tools capable of producing art, stories, code, and media. The global generative AI content creation market, valued at USD 14.8 billion in 2024, is expected to reach USD 80.12 billion by 2030, growing at a compound annual growth rate (CAGR) of 32.5% from 2025 to 2030.

But with the rise of AI content, the lawsuits against companies building such AI models have also been flocking the courts. Most of these lawsuits centers around copyright infringement and data breaches that is caused when these AI models are trained on large amount of copyrighted data.

This also poses a fundamental challenge to traditional concepts of creativity and ownership. It forces us to re-examine what it means to be an author, and how societies can balance innovation with fairness in the digital age.

Understanding the Basics: What is Copyright?

Before diving into how AI challenges copyright law, it’s essential to understand what copyright actually protects. Copyright is a form of intellectual property (IP) that grants the creator of an original work exclusive rights to use, reproduce, distribute, and display that work. It applies to artistic, literary, and musical works, among others.[1]

The cornerstone of copyright law lies in human creativity—the expression of an individual’s intellect and emotions. Traditional copyright systems across the world, including the U.S. Copyright Act of 1976, the Berne Convention, and the European Union’s Copyright Directive, all emphasise the requirement of human authorship.[2]

This creates a major conflict when applied to artificial intelligence systems, which can generate creative outputs without direct human involvement. The growing capabilities of AI such as generating artwork using DALL·E, composing music with AIVA, or writing stories using ChatGPT , are testing the limits of traditional copyright frameworks.

The Rise of AI-Generated Creativity

AI systems today are capable of producing works that rival or, in some cases, even surpass human creativity. They use deep learning models[3] trained on massive datasets of existing human-created works to learn patterns, structures, and styles. Once trained, these models can generate new, original content ranging from paintings and poems to software code and cinematic scripts.

Examples include:

DALL·E and Midjourney: AI models that generate high-quality artwork and designs from text prompts.
AIVA and Amper Music: Systems that compose music across multiple genres.
ChatGPT and Jasper AI: Language models capable of writing articles, books, and even legal documents.
DeepMind’s AlphaFold: A model that creates complex biological structure predictions, blending science and creativity.

These developments have blurred the lines between human and machine creativity, raising profound legal questions in the realm of AI and copyright.

AI Training Data and Copyright Infringement

Another crucial aspect of the AI and copyright debate concerns how AI learns. Most AI systems are trained using massive datasets scraped from the internet, which often include copyrighted works such as books, songs, photos, and videos without explicit permission.

The question arises: Does using copyrighted material for AI training constitute infringement?

There are two competing perspectives:

Fair Use Doctrine (U.S. Context): Some argue that using copyrighted content for AI training qualifies as “fair use,” especially if it transforms the data for a new purpose, such as teaching a machine learning model.
Infringement Perspective: Others argue that reproducing or storing copyrighted material in datasets violates the rights of original creators, particularly when AI-generated content competes with human-made works in the market.

The outcome of ongoing legal battles over training data will have massive implications for the future of AI and copyright law.

Who Owns AI-Generated Works?

The other pressing question in the AI and copyright debate is authorship. Since copyright law is designed to protect human creators, it’s unclear whether AI can be considered an author or if ownership should belong to its human operators.

There are three major schools of thought on this issue:

1. AI Cannot Be an Author

Most legal systems currently hold that AI cannot own copyright because it lacks human consciousness, intent, and creativity. This position aligns with the view that authorship requires human intellect and moral rights that machines cannot possess.

For example:

The U.S. Copyright Office explicitly denies registration for works “produced by a machine or mere mechanical process that operates randomly or automatically without any creative input or intervention from a human author.”
UK and EU law also emphasise human authorship. However, the UK’s Copyright, Designs and Patents Act 1988 provides a slight exception, stating that for computer-generated works with no human author, “the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.”

2. The Programmer or Developer as the Author

Another viewpoint suggests that the creators of the AI system—the programmers—should hold copyright, as they design the algorithms and models that make creative outputs possible. However, this argument faces challenges since developers may not directly control or predict the exact output of AI systems.

3. The User or Operator as the Author

A third approach attributes authorship to the user who provides the input or prompt that leads to the AI-generated work. This approach emphasises the user’s creative control and intent. However, the level of human contribution can vary widely, making this approach difficult to standardise.

The lack of consensus across jurisdictions makes AI and copyright a complex and evolving legal frontier.

Landmark Cases and Legal Precedents

Several landmark cases around the world have shaped the ongoing debate around AI and copyright:

1. Thaler v. Copyright Office (2022) – The “Creativity Machine” Case

Dr Stephen Thaler filed for copyright protection in the U.S. for an artwork titled “A Recent Entrance to Paradise,” claiming it was autonomously generated by his AI system called the Creativity Machine.
The U.S. Copyright Office rejected the claim, citing that only human authors can be recognised under the Copyright Act. Thaler appealed, but the court reaffirmed the decision in 2023, setting a major precedent that AI-generated works cannot be copyrighted in the U.S. without human authorship.

2. Getty Images vs. Stability AI (2023)

Getty Images sued Stability AI, the creator of the AI art generator Stable Diffusion, for allegedly using millions of copyrighted images without permission to train its model. This case highlighted another critical aspect of AI and copyright—the legality of using copyrighted material as training data for AI systems.

3. Sarah Silverman vs. OpenAI and Meta (2023)

Authors and artists, including comedian Sarah Silverman, filed lawsuits against AI companies claiming that their copyrighted works were used without consent to train AI models. These cases underscore the growing tension between creators and AI developers over data usage and intellectual property rights.

These precedents reflect a global struggle to adapt existing copyright frameworks to a world where machines can create, learn, and innovate.

Ethical and Philosophical Dimensions

Beyond legal disputes, AI and copyright also raise profound ethical questions about the nature of creativity and ownership.

1. Can Machines Be Creative?

True creativity is often associated with human emotions, intuition, and consciousness. AI, on the other hand, generates works based on patterns in data—it doesn’t experience inspiration or intention. Critics argue that AI merely imitates human creativity, while proponents claim that the originality of AI outputs should be recognised, regardless of the creator’s nature.

2. The Value of Human Artistry

If AI-generated works flood the market, will they devalue human-created art? Some fear that creative professions could be marginalised as machines become faster and cheaper creators. The ethical challenge lies in ensuring that technology enhances rather than replaces human creativity.

3. Accountability and Bias

Since AI systems learn from existing data, they may inadvertently replicate societal biases or stereotypes. Assigning authorship and accountability becomes difficult when AI-generated works cause harm, spread misinformation, or infringe on rights.

Global Perspectives on AI and Copyright

Different countries are approaching the issue of AI and copyright in diverse ways:

United States

The U.S. maintains a firm stance that only human authorship qualifies for copyright protection. However, the Copyright Office has released guidance stating that works involving both human and AI contributions may be partially protected if the human input is substantial and creative.

United Kingdom

The UK is one of the few countries with specific provisions for computer-generated works, granting authorship to the person who made the necessary arrangements for creation. Still, debates continue about how this applies to modern AI tools like generative models.

European Union

The EU follows a human-centric approach but has introduced AI-specific regulations such as the EU AI Act, aimed at ensuring transparency and accountability in AI development. Copyright reform discussions are ongoing to adapt to the digital era.

China

China is actively exploring frameworks to recognise AI-generated works. In 2019, a Chinese court acknowledged copyright protection for an AI-generated financial report, citing significant human involvement in the process.

Japan

Japan allows the use of copyrighted material for AI training without the rights holder’s consent, provided it’s for information analysis—a move aimed at promoting AI innovation.

These differing national approaches reveal the lack of a unified global framework for AI and copyright, potentially leading to cross-border conflicts and inconsistencies.

Possible Solutions and Future Directions

As the debate intensifies, experts and lawmakers are proposing several solutions to reconcile AI and copyright issues:

1. New Legal Frameworks

Some suggest creating new categories of intellectual property specifically for AI-generated works. These could grant limited rights to developers or users without undermining the traditional notion of human authorship.

2. Shared Authorship Models

Another approach is recognising shared authorship between humans and AI systems, acknowledging both the human input and the algorithm’s role in creation.

3. Licensing and Data Transparency

AI developers could be required to obtain licenses or maintain transparency about the datasets used for training. This would ensure fair compensation to original creators and promote ethical data usage.

4. Moral Rights and Human Oversight

Even if AI-generated works gain protection, humans should retain moral rights—ensuring accountability, integrity, and ethical oversight of AI-driven creativity.

5. International Harmonisation

Given the global nature of AI technology, international cooperation is essential. Global treaties or standards, possibly under organisations like the World Intellectual Property Organisation (WIPO), could help harmonise approaches to AI and copyright.

The Road Ahead: Balancing Innovation and Protection

The future of AI and copyright will depend on how well we balance two competing goals: encouraging innovation and protecting human creativity. If laws remain too rigid, they may stifle AI advancements. But if they become too lenient, they could undermine the rights and livelihoods of human artists and creators.

Ultimately, the path forward requires thoughtful collaboration among governments, technologists, artists, and legal scholars. Ethical AI innovation must coexist with strong intellectual property rights to ensure a fair, creative, and inclusive digital future.

Conclusion

The debate over AI and copyright represents one of the most transformative legal challenges of the 21st century. It forces society to confront fundamental questions about creativity, ownership, and the role of technology in art and innovation.

While current laws largely exclude AI from copyright protection, the rapid pace of technological progress demands adaptive and forward-thinking solutions. Whether through new legislation, hybrid authorship models, or international cooperation, the world must redefine how we recognise and reward creativity in an age where machines are not just tools—but collaborators.

As we stand at the intersection of human ingenuity and artificial intelligence, the resolution of the AI and copyright debate will shape not only the future of law but also the very essence of what it means to be a creator.