Legal aspects of using synthetic data
DOI:
https://doi.org/10.17308/law/1995-5502/2026/2/100-111Keywords:
data, data synthesis, synthetic data, personal data, artificial intelligence, training data, big dataAbstract
Society increasingly relies on generative artificial intelligence models, which are used in many areas and have become an integral part of the lives of various social groups. The exponential growth of data, however, cannot meet the ever-growing demand for training artificial intelligence models. A relatively new way to overcome the shortage of training data is the use of synthetic data. Synthetic data is used in various fields, including medicine, agriculture, industry, and urban planning, making it relevant to study the legal implications of their active use. The aim of this article is to examine the legal aspects of the use of synthetic data. The objectives of the article are to examine issues related to the concept of synthetic data, ensuring confidentiality when using synthetic data, risks in this area, including the spread of bias, the degradation of artificial intelligence models, disinformation, and international and foreign approaches to regulating the use of synthetic data. In writing this article, such methods as comparative legal, formal legal, analysis and synthesis, deduction and induction, and legal modeling were used.
Downloads
References
Baturin Yu. M. Will network media fall into the artificial intelligence trap? : speech at the XXXIII International Scientific-Practical Conference «Mediapravo – 2024» (Moscow, December 12–13, 2024).
Bryukhovetsky I. V. () Legal features of labeling content created by generative artificial intelligence in China // Bulletin of the University named after O. E. Kutafi n (MSAL). 2024. No. 10. P. 117–125.
Minbaleev A. V., Storozhakova E. E. Problems of legal protection of personal data in the process of using neural networks // Bulletin of the University named after O. E. Kutafi n (MSAL). 2023. No. 2. P. 71–79.
Savelev A. I. Towards a data regulation concept in the digital economy // Law. 2019. No. 4. P. 174–195.
D'Amico S., Dall'Olio D., Sala C., et al. Synthetic data generation by artificial intelligence to accelerate research and precision medicine in hematology // JCO Clin Cancer Inform. 2023. Jun. URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10569771/
Emam K. El., Mosquera L., Bass J. Evaluating identity disclosure risk in fully synthetic health data : Model development and validation // J Med Internet Res. 2020. Vol. 22. No. 11. URL: https://www.jmir.org/2020/11/e23139/
Gerstgrasser Matthias, Schaeffer Rylan, Dey Apratim, Rafailov Rafael, Sleight Henry, Hughes John, Korbak Tomasz, Agrawal Rajashree, Pai Dhruv, Gromov Andrey, Roberts Daniel A., Yang Diyi, Donoho David L., Koyejo Sanmi. Is model collapse inevitable? Breaking the curse of recursion by accumulating real and synthetic data. 2024. URL: https://arxiv.org/abs/2404.01413
James Jordon, Lukasz Szpruch, Florimond Houssiau, Mirko Bottarelli, Giovanni Cherubin, Carsten Maple, Samuel N. Cohen, Adrian Weller. Synthetic Data – what, why and how? URL: https://royalsociety.org/-/media/policy/projects/privacy-enhancing-technologies/ Synthetic_Data_Survey-24.pdf
Kazdan Joshua, Schaeffer Rylan, Dey Apratim, Gerstgrasser Matthias, Rafailov Rafael, Donoho David L., Koyejo Sanmi. Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World. 2024. URL: https://arxiv.org/abs/2410.16713
Khaled El Emam. Status of synthetic data generation for structured health data // JCO Clinical Cancer Informatics. 2023. Vol. 7. URL: https://ascopubs.org/doi/full/10.1200/CCI.23.00071
Myles P., Ordish J., Tucker A. The potential synergies between synthetic data and in silico trials in relation to generating representative virtual population cohorts // Progress in Biomedical Engineering. 2023. No. 5. URL: https://iopscience.iop.org/article/10.1088/2516-1091/acafbf/meta
Offenhuber D. Shapes and frictions of synthetic data // Big Data & Society. 2024. 11 (2). URL: https://www.researchgate.net/publication/380288535_Shapes_and_frictions_of_synthetic_data
Ohm P. Broken promises of privacy : responding to the surprising failure of anonymization // UCLA Law Review. 2010. No. 57. P. 1701–1777.
Pablo Villalobos, Anson Ho, Jaime Sevilla, Tamay Besiroglu, Lennart Heim, Marius Hobbhahn. Will we run out of data? Limits of LLM scaling based on human-generated data. URL: https://arxiv.org/abs/2211.04325
Shumailov Ilia, Shumaylov Zakhar, Zhao Yiren, Gal Yarin, Papernot Nicolas, Anderson Ross. The curse of recursion : training on generated data makes models forget. 2023. URL: https://arxiv.org/abs/2305.17493
