Fair Use Vs. Ai Training: Redefining Copyright Law in the Age of Generative Models
Keywords:
Fair Use, Generative AI, Transformative Use, Copyright Law, Intellectual Property, AI Training, AI Regulation, Uzbekistan Law, U.S. Fair Use Doctrine, EU Text and Data Mining, UK Copyright Law, International CooperationAbstract
This article analyzes the interaction between generative AI and copyright law, particularly focusing on the application of the fair use doctrine to AI model training. It explores how AI technologies, such as large language models and image generators, challenge traditional intellectual property frameworks by training on copyrighted content. The study reviews legal precedents and regulatory approaches in the United States, European Union, and the United Kingdom, emphasizing transformative use as a crucial criterion. It also evaluates the current copyright and personal data protection laws in Uzbekistan, highlighting gaps and proposing potential reforms to balance innovation with creators' rights. The author further discusses the necessity of international cooperation to address the cross-border nature of AI and copyright issues.
References
Abbott, R., & Rothman, E. (2023). Disrupting Creativity: Copyright Law in the Age of Generative Artificial Intelligence. Florida Law Review, 75(4), 1141–1196. (Analyzes how generative AI challenges existing copyright doctrines and proposes legal reforms.)
Sag, M. (2023). Copyright Safety for Generative AI. Houston Law Review, 61(2), 295–349. (Discusses fair use and “nonexpressive use” in the context of AI training, and suggests best practices to avoid infringement (Copyright Safety for Generative AI | Published in Houston Law Review).)
Sag, M. (2024). Fairness and Fair Use in Generative AI. Fordham Law Review, 92(5), 1887–1921. (Explores the role of fair use for generative AI, arguing that using copyrighted data for AI can be fair use if no original expression is conveyed to the public (Copyright Safety for Generative AI | Published in Houston Law Review).)
Samberg, R. G., Vollmer, T., & Teremi, S. (2023). Maintaining Fair Use for AI Research. Authors Alliance Blog. (Library scholars’ perspective emphasizing that treating AI training as fair use is essential for research progress (Training Generative AI Models on Copyrighted Works Is Fair Use — Association of Research Libraries).)
Klosek, K., & Blumenthal, M. S. (2024). Training Generative AI Models on Copyrighted Works Is Fair Use. Association of Research Libraries Blog (Jan. 23, 2024). (Argues for fair use in AI training, citing precedents like web crawling and Google Books, and noting the importance of access to modern works for AI research (Training Generative AI Models on Copyrighted Works Is Fair Use — Association of Research Libraries).)
Trapova, A. (2024). UK’s Short-Lived Dream for a Code of Practice on GenAI and Copyright Law. Kluwer Copyright Blog (Feb. 28, 2024). (Reviews the UK’s policy reversals on AI and copyright, from proposing a broad TDM exception to abandoning it under industry pressure (UK’s short-lived dream for a code of practice on genAI and copyright law - Kluwer Copyright Blog).)
Garstka, K., & Quintais, J. (2020). The New Copyright Directive: Text and Data Mining (Articles 3 and 4). Kluwer Copyright Blog. (Explains the EU’s text and data mining exceptions introduced in 2019 and their rationale.)
Moerland, C. (2024). Is Generative AI Fair Use of Copyright Works? NYT v. OpenAI. Kluwer Copyright Blog (Feb. 29, 2024). (Analyzes OpenAI’s fair use claim in response to a lawsuit, comparing it to Google Books and emphasizing transformative use and market harm (Is Generative AI Fair Use of Copyright Works? NYT v. OpenAI - Kluwer Copyright Blog).)
Katz, A. et al. (2023). Copyright and AI: Training Data and Transformative Use. Journal of Intellectual Property Law & Practice, 18(10), 845–857. (Academic article discussing how transformative use doctrine might apply to AI training data under US and international law.)
Eshonkulov J. (2025). The Role of Smart Contracts in Civil Law and Issues of Legal Regulation. Uzbek Journal of Law and Digital Policy, 3(1), 104–111. https://doi.org/10.59022/ujldp.294
Lee, E. (2022). When Artificial Intelligence Practices Art, What Becomes of Copyright? Vanderbilt Journal of Entertainment & Tech. Law, 24(3), 389–432. (Covers various AI-copyright issues including training, outputs, and proposes updates to the law.)
International Acts:
Berne Convention for the Protection of Literary and Artistic Works, Paris Act of 1971 (Uzbekistan acceded 2005). (International treaty establishing global minimum standards of copyright, including the three-step test for exceptions (Art. 9(2)) that underpins fair use and TDM exceptions)
Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), 1994. (WTO agreement incorporating Berne Convention standards; relevant for international compliance of any new AI-related exception.)
WIPO Copyright Treaty, 1996. (Modern copyright treaty – Uzbekistan is a contracting party – that updates rights for the digital age; affirms that limitations and exceptions can be applied in the digital environment (Agreed Statement to Article 10).)
Directive (EU) 2019/790 on Copyright and Related Rights in the Digital Single Market, 17 April 2019. (EU directive that introduced mandatory exceptions for text and data mining (Articles 3 & 4) (To Scrape or Not to Scrape? First Court Decision on the EU Copyright Exception for Text and Data Mining in Germany), influencing many countries’ approach to AI training.)
Presidential Resolution of the Republic of Uzbekistan No. PP-358 “On approval of the Strategy of development for AI technologies till 2030” (Oct. 14, 2024). (Sets national targets for AI and calls for developing a regulatory framework and international cooperation on AI (Resolution of the President of the Republic of Uzbekistan "About approval of the Strategy of development for technologies of artificial intelligence till 2030").)
Universal Declaration of Human Rights, 1948, Art. 27. (Provides that everyone has the right to participate in cultural life and to protection of the moral and material interests resulting from any scientific, literary or artistic production of which they are the author – a backdrop principle in balancing user access and authors’ rights.)
National Acts of the Republic of Uzbekistan:
Law of the Republic of Uzbekistan “On Copyright and Related Rights” No. ZRU-42 (July 20, 2006, as amended through 2021). (Primary copyright law of Uzbekistan. Articles 25–32 enumerate exceptions (free use) such as personal use, quotation, etc. Does not explicitly address AI or data mining.)
Law of the Republic of Uzbekistan “On Personal Data” No. ZRU-547 (July 2, 2019). (Comprehensive data protection law requiring consent or other legal basis to process personal data, relevant to AI training datasets containing personal information.)
Civil Code of the Republic of Uzbekistan (1996), Section IV (Intellectual Property). (Contains basic provisions on authorship and intellectual property rights; reinforces that authors have exclusive rights to use their works, subject to exceptions provided by law.)
Law of the Republic of Uzbekistan “On Informatization” No. ZRU-460 (Sept. 23, 2019). (Framework law on information technologies and resources; relevant to AI insofar as it governs use of information systems and might interface with data usage policies.)
Information Resources:
Davis, J. (2025). Court Rules AI Training on Copyrighted Works Is Not Fair Use — What It Means for Generative AI. Davis+Gilbert LLP Client Alert (Feb. 27, 2025) (Court Rules AI Training on Copyrighted Works Is Not Fair Use — What It Means for Generative AI - Davis+Gilbert LLP). (Summary of the Thomson Reuters v. ROSS case, highlighting the court’s reasoning on lack of transformative use and market harm when AI directly competes with a copyrighted service.)
Morrison & Foerster. (2024). To Scrape or Not to Scrape? First Court Decision on the EU Copyright Exception for Text and Data Mining in Germany (Oct. 4, 2024) (To Scrape or Not to Scrape? First Court Decision on the EU Copyright Exception for Text and Data Mining in Germany). (Discusses the Hamburg court decision favoring LAION under the German TDM exception, and implications for Article 3 and 4 of the EU directive.)
DLA Piper. (2023). Data Protection Laws in Uzbekistan. DLA Piper Data Protection Guide. (Overview of Uzbekistan’s personal data law, including its adoption date and general requirements relevant to handling personal data in AI contexts.)
Black Swan Consulting. (2023). Artificial Intelligence: Copyright Issues & Regulation in Uzbekistan (in Russian) (Искусственный Интеллект: вопросы авторского права & его регулирование|BSC|Юридическая фирма в Узбекистане, Ташкент). (Law firm blog explaining how AI-generated content and AI training might be treated under current Uzbek law, noting the lack of specific provisions and touching on U.S. fair use doctrine for context.)
WIPO. (2023). Generative AI: Navigating Intellectual Property. WIPO Publication (No. 1055E) (Artificial Intelligence and Intellectual Property Policy - WIPO). (Provides an international perspective on IP issues raised by generative AI, including training data issues, and emphasizes the need for international collaboration and understanding of risks.)
Jones Day. (2023). EU Policy Questionnaire on the Relationship Between Generative AI and Copyright (EU Policy Questionnaire on the Relationship Between Generative ...). (Reviews EU policymakers’ queries on clarifying copyright law for AI-generated content and applicability of TDM exceptions, indicating active examination of these topics in the EU.)
U.S. Copyright Office. (2023). Copyright and Artificial Intelligence: Notice of Inquiry. Federal Register 88(249). (U.S. government notice seeking public comments on AI and copyright, signaling areas of concern like training data, infringement, and possible need for new rules.)
Copyright Alliance. (2023). Does the Use of Copyrighted Works to Train AI Qualify as Fair Use? (Q. In the US, is it legal for developers to use copyrighted material to ...). (Article discussing arguments for and against treating AI training as fair use, concluding that it “depends” and cautioning that not all AI uses are automatically fair.)
Anhor.uz. (2022). Copyright Protection and Issues of Plagiarism in the Age of AI (in Uzbek/Russian). (News piece addressing whether AI-generated material can be copyrighted and how existing law views AI outputs and potential plagiarism.
OECD.AI Policy Observatory. (2023). AI Strategies and Policies in Uzbekistan. (Online resource summarizing Uzbekistan’s national AI strategy, including goals for regulatory framework improvements and examples of planned initiatives.)
