Algorithmic Crawling vs. Copyright Protection: Defining the Boundaries of Fair Use in Text and Data Mining for AI Training
European Journal of Risk Regulation · 08.07.2026 · 1-18
DOI: 10.1017/err.2026.10118 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Journal of Risk Regulation, online erschienen am 08.07.2026. Verfasst von Siyang Chen und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The advancement of AI relies on text and data mining (TDM) to acquire training data, yet its large-scale algorithmic crawling activities fundamentally conflict with copyright regimes. Through comparative legal research, this article reveals the divergent governance approaches and efficacy deficiencies within the three major jurisdictions of the European Union, the United States and China. We argue that the current regulatory model adopted by the EU may adversely affect the dynamism of Europe’s AI industry. At the policy level, the EU should construct a synergistic “Cost-Benefit-Governance” framework, reducing compliance costs through differentiated regulation, collective licensing, and fiscal support measures. It should also leverage unified internal rules and international multilateral platforms to foster governance consensus. In legal practice, the EU should advance the clarification and tiered application of copyright exceptions. This can be achieved by refining rules, implementing data classification governance and innovating safe harbour liability mechanisms to enhance legal predictability, thereby balancing technological innovation and copyright protection. This framework aims to mitigate the inhibitory effect of current systems on AI innovation and address jurisdictional barriers through international cooperation.
Zitieren
APA
Chen, S., & Long, X. (2026). Algorithmic Crawling vs. Copyright Protection: Defining the Boundaries of Fair Use in Text and Data Mining for AI Training. European Journal of Risk Regulation, 1-18. https://doi.org/10.1017/err.2026.10118
Vancouver
Chen S, Long X. Algorithmic Crawling vs. Copyright Protection: Defining the Boundaries of Fair Use in Text and Data Mining for AI Training. European Journal of Risk Regulation. 2026:1-18. doi:10.1017/err.2026.10118
BibTeX
@article{Chen2026,
author = {Chen, Siyang and Long, Xinxuan},
title = {Algorithmic Crawling vs. Copyright Protection: Defining the Boundaries of Fair Use in Text and Data Mining for AI Training},
journal = {European Journal of Risk Regulation},
year = {2026},
pages = {1--18},
doi = {10.1017/err.2026.10118}
}
EUVIMED