Copyright in generative deep learning
Open accessThe DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Machine-generated artworks are now part of the contemporary art scene: they are attracting significant investments and they are presented in exhibitions together with those created by human artists. These artworks are mainly based on generative deep learning (GDL) techniques, which have seen a formidable development and remarkable refinement in the very recent years. Given the inherent characteristics of these techniques, a series of novel legal problems arise. In this article, we consider a set of key questions in the area of GDL for the arts, including the following: is it possible to use copyrighted works as training set for generative models? How do we legally store their copies in order to perform the training process? Who (if someone) will own the copyright on the generated data? We try to answer these questions considering the law in force in both the United States and the European Union, and potential future alternatives. We then extend our analysis to code generation, which is an emerging area of GDL. Finally, we also formulate a set of practical guidelines for artists and developers working on deep learning generated art, as well as some policy suggestions for policymakers.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
A Style-Based Generator Architecture for Generative Adversarial Networks
Karras, T., Laine, S., & Aila, T. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00453 · Cited by 11,714
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557
WaveNet: a generative model for raw audio
Oord, A. V. D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1609.03499 · Cited by 4,911
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856
Large Scale GAN Training for High Fidelity Natural Image Synthesis
Brock, A., Donahue, J., & Simonyan, K. (2018) · arXiv · doi:10.48550/arxiv.1809.11096 · Cited by 3,717
Self-Attention Generative Adversarial Networks
Zhang, H., Goodfellow, I., Metaxas, D., & Odena, A. (2018) · arXiv · doi:10.48550/arxiv.1805.08318 · Cited by 3,353
Taming Transformers for High-Resolution Image Synthesis
Esser, P., Rombach, R., & Ommer, B. (2020) · arXiv · doi:10.48550/arxiv.2012.09841 · Cited by 2,514
Research works printed but not linked: 99
- Samuelson, 2009, Unbundling fair uses, Fordham Law Review, 77, 2537
- Foster, 2019, Generative Deep Learning
- Chiou, 2019, Copyright lessons on machine learning: What impact on algorithmic art, Journal of Intellectual Property, Information Technology and Electronic Commerce Law, 10, 398
- Ligon, SE (2019) AI can create art, but can it own copyright in It, or infringe? Available at lexisnexis.com/practicalguidance/the-journal/b/pa/posts/ai-can-create-art-but-can-it-own-copyright-in-it-or-infringe. Retrieved June 18, 2021.
- Colton, 2008, AAAI 2008 Spring Symposium
- Wu, B , Zhao, S , Chen, C , Xu, H , Wang, L , Zhang, X , Sun, G and Zhou, J (2019) Generalization in generative adversarial networks: A novel perspective from privacy protection. In NeurIPS. Vancouver, Canada: Curran Associates, Inc., Vol. 32, pp. 307–317.
- Ginsburg, 2014, Fair use for free, or permitted-but-paid?, Berkeley Technology Law Journal, 29, 1383
- Bohlen, HM (2017) EU Copyright Protection of Works Created by Artificial Intelligence Systems. Master’s thesis. Bergen, Norway: University of Bergen.
- Denicola, 2016, Ex machina: Copyright protection for computer-generated works, Rutgers Law Review, 69, 251
- 10.2139/ssrn.3299523 · DOI
- 10.3389/fdigh.2017.00003 · DOI
- Hurt, 1966, The economic rationale of copyright, The American Economic Review, 56, 421
- Franceschelli, G and Musolesi, M (2021a) Copyright in generative deep learning. Preprint, arXiv:2105.09266 [cs.CY]. · arXiv
- Guadamuz, 2017, Do androids dream of electric copyright? Comparative analysis of originality in artificial intelligence generated works, Intellectual Property Quarterly, 2, 169
- Jaques, 2016, NeurIPS
- 10.2307/1341682 · doi:10.2307/1341682
- Nimmer, 2003, “Fairest of them all” and other fairy Tales of fair use, Law and Contemporary Problems, 66, 263
- Zhang, 2014, EMNLP, 670
- Sag, 2019, The new legal landscape for text mining and machine learning, Journal of the Copyright Society of the USA, 66, 291
- Isola, 2017, CVPR, 5967
- 10.1007/s00521-020-05383-8 · DOI · Paper page
- Farr, 1989, Copyrightability of computer-created works, Rutgers Computer and Technology Law Journal, 15, 63
- Lessig, 2004, Free Culture: How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity
- Carroll, 2007, Fixing fair use, North Carolina Law Review, 85, 68
- 10.2139/ssrn.2900110 · doi:10.2139/ssrn.2900110
- Palace, 2019, What if artificial intelligence wrote this? Artificial intelligence and copyright law, Florida Law Review, 71, 217
- Tilson, 2019, CCECE, 1
- Netanel, 2011, Making sense of fair use, Lewis and Clark Law Review, 15, 715
- 10.22161/ijaers.712.9 · doi:10.22161/ijaers.712.9
- Xu, 2018, Big Data, 570, 10.1007/978-981-13-2922-7 · doi:10.1007/978-981-13-2922-7
- Goodfellow, 2016, Deep Learning
- Liu, 2019, ICPADS
- Ramesh, 2021, ICML, 139, 8821
- Potash, 2015, EMNLP, 1919
- Van Den Oord, 2016a, ISCA. Speech Synthesis Workshop, 125
- 10.1016/j.patter.2020.100026 · doi:10.1016/j.patter.2020.100026
- Kingma, 2014, ICLR
- Gervais, 2002, Feist goes global: A comparative analysis of the notion of originality in copyright law, Journal of the Copyright Society of the U.S.A., 49, 949
- Ralston, 2004, Copyright in computer-composed music: HAL meets Handel, Journal of the Copyright Society of the U.S.A., 52, 281
- Butler, 1982, Can a computer be an author—Copyright aspects of artificial intelligence, Hastings Communications and Entertainment Law Journal, 4, 707
- Samuelson, 1986, Symposium Cosponsored by University of Pittsburgh law Review and the Software En on the Future of Software Protection, 1185
- Devlin, 2019, NAACL 2019, 4171
- 10.1080/10192557.2019.1705525 · DOI · Paper page
- 10.1093/acprof:oso/9780195137620.001.0001 · DOI
- Sturm, 2019, Artificial intelligence and music: Open questions of copyright law and engineering praxis, The Artist and Journal of Home Culture, 8, 115
- 10.2139/ssrn.3160586 · DOI · Paper page
- Yi, 2018, EMNLP, 3143
- 10.1109/TPAMI.2013.50 · DOI · Paper page
- 10.5771/2568-9185-2020-2-390 · DOI
- 10.1145/3376898 · DOI · Paper page
- 10.1007/s40319-018-0722-2 · DOI
- Yu, 2020, ECCV, 12367
- Glasser, 2001, Copyrights in computer-generated works: Whom, if anyone, do we reward?, Duke Law and Technology Review, 24
- Zugarini, 2019, ICANN, 313
- Boyden, 2016, Emergent works, The Columbia Journal of Law and the Arts, 3, 377
- Gurkaynak, 2017, Questions of intellectual property in the artificial intelligence realm, The Robotics Law Journal, 3, 9
- 10.1007/s40319-018-0670-x · DOI · Paper page
- Gervais, 2020, The machine as author, Iowa Law Review, 105, 1053
- Hristov, 2017, Artificial intelligence and the copyright dilemma, IDEA: The IP Law Review, 57, 24
- Lau, 2018, ACL, 1948
- Bonadio, 2020, Artificial intelligence as producer and consumer of copyright works: Evaluating the consequences of algorithmic creativity, Intellectual Property Quarterly, 2, 112
- Sturm, BL , Santos, JF , Ben-Tal, O and Korshunova, I (2016) Music transcription modelling and composition using deep learning. Preprint, arXiv:1604.08723 [cs.SD]. · arXiv
- Yu, 2017, AAAI, 2852
- Grimmelmann, 2016b, There’s no such thing as a computer-authored work—And It’s a good thing, too, Columbia Journal of Law and the Arts, 39, 403
- 10.2139/ssrn.3261329 · DOI
- Goodfellow, 2014, NeurIPS, 2672
- Bolukbasi, 2016, NeurIPS, 4356
- Wu, 1997, From video games to artificial intelligence: Assigning copyright ownership to works generated by increasingly sophisticated computer programs, AIPLA Quarterly Journal, 25, 131
- Schonberger, 2018, Droit d’auteur 4.0/Copyright 4.0, 145
- Bridy, 2012, Coding creativity: Copyright and the artificially intelligent author, Stanford Technology Law Review, 5, 1
- Esser, 2021, CVPR, 12868
- Geiger, 2015, The resolution of the European Parliament of 9 July 2015: Paving the way (finally) for a copyright reform in the European Union?, European Intellectual Property Review, 37, 683
- 10.1093/jiplp/jpz056 · doi:10.1093/jiplp/jpz056
- Sobel, 2020, Artificial Intelligence and Intellectual Property
- Brown, 2020, NeurIPS, 33, 1877
- Lunney, 2002, Fair use and market failure: Sony revisited, Boston University Law Review, 82, 975
- Karras, 2018, ICLR
- Hadjeres, 2017, ICML, 70, 1362
- 10.1016/j.eswa.2015.01.041 · DOI · Paper page
- Van Den Oord, 2016c, NeurIPS, 4797
- Asay, 2020, Is transformative use eating the world?, Boston College Law Review, 61, 905
- Grimmelmann, 2016a, Copyright for literate robots, Iowa Law Review, 101, 657
- Radford, 2021, ICML
- 10.2139/ssrn.3374951 · DOI
- Engel, 2019, ICLR: OpenReview
- 10.2139/ssrn.2957722 · DOI
- Jin, 2017, NeurIPS
- Franceschelli, 2019, I, Artist. Opere d’arte e Intelligenza Artificiale. Il Curioso Caso del Diritto d’autore
- Brock, 2019, ICLR
- Karras, 2019, CVPR, 4396
- Huang, 2019, ICLR
- Elgammal, 2017, ICCC
- Gregor, 2015, ICML., 37, 1462
- Franceschelli, G and Musolesi, M (2021b) Creativity and machine learning: A survey. arXiv:2104.02726 [cs.LG]. · arXiv
- 10.1109/TIFS.2019.2897874 · DOI · Paper page
- Van Den Oord, 2016b, ICML, 1747
- Dong, H-W , Hsiao, W-Y , Yang, L-C and Yang, Y-H (2018) MuseGAN: Multi-track sequential generative adversarial networks for symbolic music generation and accompaniment. In AAAI. New Orleans, LA: AAAI Press, Vol. 32(1), pp. 34–41. · DOI · Paper page
- Radford, 2019, Language models are unsupervised multitask learners, OpenAI Blog, 1, 9
- Beaulieu-Jones, 2019, Privacy-preserving generative deep neural networks support clinical data sharing, Circulation: Cardiovascular Quality and Outcomes, 12, e005122
Cited by
Draxler, F., Werner, A., Lehmann, F., Hoppe, M., Schmidt, A., Buschek, D., & Welsch, R. (2023) · arXiv · doi:10.48550/arxiv.2303.03283 · Cited by 284
Privacy and Security Concerns in Generative AI: A Comprehensive Survey
Golda, A., Mekonen, K., Pandey, A., Singh, A., Hassija, V., Chamola, V., & Sikdar, B. (2024) · IEEE Access · doi:10.1109/access.2024.3381611 · Cited by 264
Best humans still outperform artificial intelligence in a creative divergent thinking task
Koivisto, M., & Grassini, S. (2023) · Scientific Reports · doi:10.1038/s41598-023-40858-3 · Cited by 247
Design Principles for Generative AI Applications
Weisz, J. D., He, J., Muller, M., Hoefer, G., Miles, R., & Geyer, W. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642466 · Cited by 232
Draxler, F., Werner, A., Lehmann, F., Hoppe, M., Schmidt, A., Buschek, D., & Welsch, R. (2024) · ACM Transactions on Computer-Human Interaction · doi:10.1145/3637875 · Cited by 186
On the creativity of large language models
Franceschelli, G., & Musolesi, M. (2025) · AI & SOCIETY · doi:10.1007/s00146-024-02127-3 · Cited by 113
Self-citationSardesai, N., Russo, P., Martin, J., & Sardesai, A. (2024) · Postgraduate Medical Journal · doi:10.1093/postmj/qgad137 · Cited by 67
Introduction to large language models (LLMs) for dementia care and research
Treder, M. S., Lee, S., & Tsvetanov, K. A. (2024) · Frontiers in Dementia · doi:10.3389/frdem.2024.1385303 · Cited by 63
“Second, LLMs are trained on a large corpus of text from a variety of sources including social media websites, often without permission of the author of the text (Franceschelli and Musolesi, 2022; Kasneci et al., 2023).”
BackgroundInfluentialFoundation Models and Fair Use
Henderson, P., Li, X., Jurafsky, D., Hashimoto, T., Lemley, M. A., & Liang, P. (2023) · doi:10.2139/ssrn.4404340 · Cited by 57
Analysis and prevention of AI-based phishing email attacks
Eze, C. S., & Shamir, L. (2024) · Electronics · doi:10.3390/electronics13101839 · Cited by 47
“That is done by using copyrighted art for training generative AI systems, that consequently generate work heavily based on the copyrighted art it was trained with (Franceschelli and Musolesi, 2022; Samuelson, 2023).”
MethodBalancing innovation and regulation in the age of generative artificial intelligence
Wang, X., & Wu, Y. C. (2024) · Journal of Information Policy · doi:10.5325/jinfopoli.14.2024.0012 · Cited by 42
Generative AI in depth: A survey of recent advances, model variants, and real-world applications
Yazdani, S., Singh, A., Saxena, N., Wang, Z., Palikhe, A., Pan, D., Pal, U., Yang, J., & Zhang, W. (2025) · Journal of Big Data · doi:10.1186/s40537-025-01247-x · Cited by 40
“The platform replicates state-of-the-art studies in a standardized manner, incorporating various architectures like DMs, autoregressive transformers, and GANs.”
Metadata
- Authors
- Franceschelli, G., & Musolesi, M.
- Venue
- Data & Policy
- Publisher
- Cambridge University Press (CUP)
- Published
- 2022
- Volume
- 4
- ISSN
- 2632-3249 (ISSN-L)
- OpenAlex
- W3162155171
- Cited by
- 81
- References
- 21
- Type
- paratext
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.