- within Technology topic(s)
- with Senior Company Executives, HR and Finance and Tax Executives
- in United Kingdom
- with readers working within the Advertising & Public Relations, Banking & Credit and Healthcare industries
1.0 INTRODUCTION
The traditional understanding of copyright protection is built on the principle that authors and rightsholders possess the exclusive power to control how their works are used, reproduced, and commercially exploited by third parties.1 Copyright has long served as a legal mechanism to safeguard creativity, ensuring that creators are adequately rewarded for their contributions while discouraging unauthorized use.
However, the rapid rise of technology, particularly the emergence of generative artificial intelligence (AI) has begun to challenge this conventional framework. AI systems are increasingly capable of producing text, images, music, sounds and other creative outputs that often rely on vast repositories of pre-existing, copyrighted material for training.2 These developments raises difficult questions: Do the existing Copyright laws anticipate or accommodate works generated by AI? Are the current legal regimes sufficient to protect the interests of creators while also fostering innovation? And perhaps most importantly, does the conventional view of copyright protection require a fundamental rethinking in order to remain relevant in an AI-driven age?
2.0. OVERVIEW OF CASES INVOLVING ARTIFICIAL INTELLIGENCE AND COPYRIGHT USE
Recently, there has been a growing wave of cases involving alleged breaches of intellectual property rights by AI platforms. The central concern is that these systems are trained on vast repositories of existing works, often scraped from the internet, and their outputs may reproduce or adapt copyrighted material without permission.3
A prominent example is The New York Times ("The times") v. OpenAI and Microsoft.4 In this case, the Times alleged that millions of its articles were used to train ChatGPT and that the platform could generate content which mimicked or reproduced portions of its journalism almost word-for-word, thus recreating its "writing style". The Times argued this amounted to large-scale copyright infringement and posed a threat to its business model, while OpenAI and Microsoft defended their practices as "fair use" within the bounds of U.S. copyright law.
Similarly, Anthropic, the developer of Claude has faced lawsuits from various rights holders, including music publishers like Universal Music Group and Concord.5 These claims are that Anthropic's models were trained on copyrighted song lyrics and texts, and that the AI can reproduce substantial parts of those works on request.
Other lawsuits have followed in the publishing and creative industries. For instance, groups of authors, including best-selling novelists, have sued OpenAI and other AI developers, alleging that their copyrighted books were copied wholesale into training datasets.6 Although many of these cases are still pending, these highlight a significant tension, as there are varying arguments on each side of the dispute. AI companies argue that training on publicly available data is essential for innovation; on the other, creators insist that their intellectual property cannot be freely exploited without permission or compensation.
3.0. HOW AI HAS CHANGED THE CONVENTIONAL UNDERSTANDING OF COPYRIGHT LAW
At the core of this discussion lies a fundamental question: can works generated through the use of artificial intelligence be entitled to copyright protection? This concern arises from the absence of human creative intent in AI-generated works, as artificial intelligence systems merely operate based on prompts and data inputs provided by users, without independent consciousness or authorship.
Across jurisdictions, the doctrine of a "modicum of creativity" serves as a threshold in determining whether a work qualifies for copyright protection. This doctrine holds that originality exists where a work reflects some degree of intellectual effort, creativity, and judgment. While the standard of creativity required is relatively low, there must still be a minimum level of human creativity for copyright to subsist. Consequently, the challenge lies in determining whether AI-generated outputs, produced through algorithms and datasets rather than human intellect, can meet this threshold, or whether protection should instead vest in the human user who supplied the prompt or exercised sufficient creative control over the process. 7
AI has introduced complexities like the above, that challenge the very foundation of traditional copyright law. Copyright was historically designed around the concept of human authorship, premised on originality, creativity, and the ability to attribute a work to a natural person.8 However, with the rise of generative AI systems capable of producing music, art, literature, film, and software code, the question arises: who is the true author of AI-generated works?
One significant shift is the blurring line of human and machine contributions. In conventional settings, authorship is tied directly to the individual who creates the work. With AI, however, the process often involves multiple actors: the developer who trained the model, the company that owns the system, the AI itself and the user who provides prompts. Courts and policymakers now grapple with whether the creative output belongs to the user, the AI developer, or remains unprotected altogether. Another area of concern lies in the originality requirement. Traditional copyright demands a modicum of creativity or proof of a sweat of the brow from a human author.9 Yet, AI systems generate outputs by learning from vast datasets of existing works, thus raising the argument that such outputs lack "true originality" or "creative intent" since they are derivative of prior human creations.
Furthermore, AI has introduced significant complexity to discussions on copyright infringement and ownership. The use of copyrighted books, music, and artworks to train AI systems has triggered lawsuits, with rights holders claiming unauthorized copying and developers arguing fair use, as outputs are not direct reproductions. These disputes often involve derivative works and IP control, especially in creative industries where AI can replicate human voices, characters, or artistic styles. On ownership, current laws do not recognize AI as an author, making joint ownership between humans and AI legally impossible. However, where a human provides substantial creative input, the user of the AI may be deemed the sole or primary author.
4.0. HOW JURISDICTIONS REGULATE AI & COPYRIGHT – A CASE STUDY ON THE EUROPEAN UNION, UNITED KINGDOM AND UNITED STATES.
As AI become widely used, Countries around the world are scrambling with how to regulate AI in a way that protects copyright while encouraging innovation. However, each countries approach varies widely.
4.1. European Union The European Union (EU), for instance has taken a proactive and structured route. In April 2021, the European Commission proposed the first EU artificial intelligence law, establishing a risk-based AI classification system.10 The EU AI Act, expected to come into effect in 2026, does not directly regulate copyright, but indirectly shapes it by classifying AI systems by risk level. A key feature is its transparency obligations, particularly around training data, which could help rights holders assess potential infringement. Generative AI tools like ChatGPT, though not classed as high-risk, must still comply with measures such as, disclosing that content was AI-generated; ensuring models are designed to avoid illegal outputs; and publishing summaries of copyrighted data used for training.
4.2. United Kingdom The United Kingdom (UK) on the other hand has adopted a more liberal approach. The government had initially opted for AI companies to have the permission to scrape and use other persons content to develop their AI, even for commercial purposes. However, after industry critcism, especially from the creative sector, the UK government withdrew a proposal to further liberalize AI commercial AI training.11
4.3. United States The United States (US), despite being the base of most major AI developers, has yet to introduce any dedicated AI-specific copyright legislation. In the absence of such laws, the judiciary has taken a leading role in shaping the conversation, with ongoing lawsuits involving OpenAI, Anthropic, and major media companies such as Disney serving as test cases for how traditional copyright rules apply in the AI context. The U.S. Copyright Office has also contributed by holding consultations and issuing guidance, clarifying that works created entirely by AI without meaningful human involvement cannot qualify for copyright protection. Furthermore, it has indicated that creators may soon be required to disclose the use of AI in their creative processes. However, while these efforts mark important steps toward clarity, a comprehensive and coherent regulatory framework is still evolving and remains under development.12
4.4. Nigeria The laws in Nigeria do not expressly address the legal issues arising from the development and use of artificial intelligence (AI), although the Copyright Act 2022 presents a more progressive framework compared to the repealed 1988 Act, particularly through its inclusion of provisions on digital piracy and online infringement.13 However, the Act remains silent on AI-generated works and the questions from the continuous use of AI in the creation of a work.
Nevertheless, regarding progress in AI regulation, the National Artificial Intelligence Strategy (2024) sets out a policy roadmap for positioning Nigeria to become a global leader in AI by leveraging the technology for socioeconomic growth, inclusion, and sustainable development.14 Additionally, the National Information Technology Development Agency (NITDA) announced its intention to develop a National Artificial Intelligence Policy (NAIP) in 2022, and invited stakeholder input.15 While NITDA has confirmed that a draft policy has been prepared, the final version is yet to be released or implemented.
CONCLUSION
As this legal landscape unfolds, three core principles are likely to shape its development. First, transparency, as AI developers may increasingly be required to disclose the datasets used to train their models, giving rights holders a clearer view of whether their works have been exploited without permission. Second, consent, as creators and publishers may push for stronger mechanisms to either authorize or prevent the use of their works in AI training. Third, accountability, as courts or legislations will need to address the question of responsibility, determining whether liability for infringing AI outputs should rest with developers, end-users, or intermediaries.
The challenge lies in ensuring that copyright protection evolves without hindering innovation. The current wave of litigation represents an early chapter in what will likely be a sustained effort to reshape the balance between intellectual property and artificial intelligence
Footnotes
1. Section 9 -13 of the Copyright Act 2022
2. Potter Clarkson, 'What Data Is Used to Train an AI, Where Does It Come From, and Who Owns It?' (2023) https://www.potterclarkson.com/insights/what-data-is-used-to-train-an-ai-where-does-it-come-from-and-who-owns-it/ accessed 2 October 2025.
3. Potter Clarkson, 'What Data Is Used to Train an AI, Where Does It Come From, and Who Owns It?' (2023) https://www.potterclarkson.com/insights/what-data-is-used-to-train-an-ai-where-does-it-come-from-and-who-owns-it/ accessed 2 October 2025.
4. Henry Ishitani, 'NYT v. OpenAI: The Times's About-Face - Harvard Law Review' (Harvard Law Review, 10 April 2024) https://harvardlawreview.org/blog/2024/04/nyt-v-openai-the-timess-about-face/ accessed 2 October 2025.
5. Emilia David, 'Universal Music Sues AI Company Anthropic for Distributing Song Lyrics' (19 October 2023) https://www.theverge.com/2023/10/19/23924100/universal-music-sue-anthropic-lyrics-copyright-katy-perry accessed 1 October 2025.
6. Section 2 of the Copyright Act, 2022
7. University of Michigan Library, 'Research Guides: Copyright Basics: Copyrightability' (2025) https://guides.lib.umich.edu/copyrightbasics/copyrightability accessed 1 October 2025.
8. Section 2 of the Copyright Act, 2022
9. University of Michigan Library, 'Research Guides: Copyright Basics: Copyrightability' (2025) https://guides.lib.umich.edu/copyrightbasics/copyrightability accessed 1 October 2025.
10. Izuchukwu Chinedo, 'AI and IP Law: What Is the Nigerian Legal Perspective? (2025) https://inventa.com/ip-news-insights/opinion/ai-and-ip-law-what nigerian-legal-perspective accessed 23 October 2025.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
