AI Training and Copyright: Should Intellectual Property Law Allow Machines to Learn?

pdf (English)
Publiée : Οκτ 1, 2024
DOI : https://doi.org/10.12681/bioeth.39041
Pedro Martins Fernandes
University of Lisbon, Lisbon, Portugal and Intern, National Commission for Bioethics & Technoethics, Greece.
Résumé

This article examines the intricate legal landscape surrounding the use of copyrighted materials in the development of artificial intelligence (AI). It explores the rise of AI and its reliance on data, emphasizing the importance of data availability for machine learning (ML) systems. The article analyzes current relevant legislation across the European Union, United States, and Japan, highlighting the legal ambiguities and constraints posed by IP rights, particularly copyright. It discusses possible new solutions, referencing the World Intellectual Property Organization's (WIPO) call for discussions on AI and IP policy. The conclusion stresses the need to balance the interests of AI developers and IP rights holders to promote technological advancement while safeguarding creativity and originality.

Article Details
  • Rubrique
  • Original Articles
Creative Commons License

Ce travail est disponible sous la licence Creative Commons Attribution 4.0 International .

Authors who publish with this journal agree to the following terms:

  • Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution  CC BY 4.0 License, which allows for immediate free access to the work and permits any user to read, download, copy, distribute, print, search, or link to the full texts of articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose. Appropriate credit must be given by citing the author(s) and the original publication in this journal.
  • Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g. post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

We encourage authors to deposit their articles, as well as data underlying the publications, in institutional and/or other appropriate subject repositories.

Bioethica permits and encourages authors to archive the final publication pdf  in institutional (e.g. the repository of the National Hellenic Research Foundation) or other appropriate subject repositories (e.g. SSOAR repository for social sciences),  in compliance with institutional and/or funder open access policies, after publication in the BIOETHICA. Authors must provide bibliographic details that credit publication in the journal, as well as related funding details (when applicable).

Lists of institutional and other subject-based academic open access repositories can be found listed by country at the registry http://opendoar.org/countrylist.php   

If your institution does not possess a repository you may deposit a copy of your paper at no cost with www.zenodo.org , the repository supported for open access research in the EU by the European Commission, through the project OpenAIRE (www.openaire.eu )

Téléchargements
Les données relatives au téléchargement ne sont pas encore disponibles.
Références
Regulation (EU) 2022/868 of the European Parliament and of the Council of 30 May 2022 on European data governance (Data Governance Act).
Levendowski A. How Copyright Law Can Fix Artificial Intelligence's Implicit Bias Problem, 93 Wash. L. Rev. 579 (2018). pp. 602 - 619. Available at: https://digitalcommons.law.uw.edu/wlr/vol93/iss2/2.
European Parliament resolution of 20 Octo-ber 2020 on intellectual property rights for the development of artificial intelligence technologies (2020/2015(INI)).
Berne Convention for the Protection of Lit-erary and Artistic Works (as amended on September 28, 1979).
Case C-05/08 Infopaq International, ECLI:EU:C:2009:465.
Case C-604/10 Football Dataco, at 39.
Case C-145/10 Painer, ECLI:EU:C:2011:798.
Directive 2001/29/EC of the European Par-liament and of the Council of 22 May 2001 on the harmonization of certain aspects of copyright and related rights in the infor-mation society.
Sousa e Silva N. ‘Inteligência Artificial e Propriedade Intelectual: Está tudo bem?’ I Congresso de Inteligência Artificial e Direito, Edições Almedina (2023), pp. 201-220.
Ducato R, Strowel A. "Limitations to Text and Data Mining and Consumer Empower-ment Making the Case for a Right to “Ma-chine Legibility”. CRIDES Working Paper Series, 31 October 2018.
Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artifi-cial intelligence (Artificial Intelligence Act).
Constitution of the United States.
Leval PN. Commentary, Toward a Fair Use Standard,103 HARV. L.REV. 1105, 1111 (1990).
Copyright Law of the United States and Re-lated Laws Contained in Title 17 of the United States Code, pp. 20.
Authors Guild, Inc. v. Google Inc., No. 13-4829-cv (2d Cir. Oct. 16, 2015).
H.R.7913 - To require a notice be submitted to the Register of Copyrights with respect to copyrighted works used in building genera-tive AI systems, and for other purposes. (Bill for Generative AI Copyright Disclosure Act)
Fukuoka S, Murata T, Mizuguchi A. Legal Issues in Generative AI under Japanese Law - Copyright. Robotics / Artificial Intelligence Newsletter, 2023 .
Basic ideas on flexible rights limitation pro-visions in response to the development of digitization and networking (related to Arti-cles 30-4, 47-4 and 47-5 of the Japanese Copyright Act), Japan Copyright Office.
WIPO, WIPO Conversation on Intellectual Property (IP) and Artificial Intelligence (Al), Draft Issues Paper on Intellectual Property and Artificial Intelligence, Second Session, WIPO Secretariat, pp. 4 and 5, available at: https://www.wipo.int/edocs/mdocs/mdocs/en/wipo_ip_ai_2_ge_20/wipo_ip_ai_2_ge_20_1.pdf.
EUROPEAN PARLIAMENT, Resolution on intellectual property rights for the devel-opment of artificial intelligence technologies (2020/2015(INI), 20.10.20, available at: https://www.europarl.europa.eu/doceo/document/TA-9-2020-0277_EN.html.
Kublik V. EU/US Copyright Law and Impli-cations on ML Training Data. Valohai, 2024. available at: https://valohai.com/blog/copyright-laws-and-machine-learning/.
Kop M. Machine Learning & EU Data Shar-ing Practices (March 3, 2020). Stanford - Vi-enna Transatlantic Technology Law Forum, Transatlantic Antitrust and IPR Develop-ments, Stanford University, Issue No. 1/2020, Available at SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3409712
https://www.japaneselawtranslation.go.jp/ja/laws/view/4207 (unofficial English translation of the Copyright Act of Japan (Act No. 48 of 1970)).
Grimmelmann J. Copyright for Literate Ro-bots. Cornell Law Faculty Publications, 2016
Joint comment to WIPO on copyright and Artificial Intelligence. Available at: https://infojustice.org/archives/42009
Lemley MA, Casey B. Fair Learning (Janu-ary 30, 2020). Available at SSRN: https://ssrn.com/abstract=3528447 or https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3528447