Artificial Intelligence Redefines the Boundaries of Intellectual Property
The academic research on the links between artificial intelligence and intellectual property law has been growing rapidly since 2019, marked by the emergence of generative models. This momentum reflects a global awareness of the legal challenges raised by these technologies, which are profoundly transforming creative and industrial practices.
Three major axes of reflection currently dominate the debates. The first concerns copyright law, where issues of attribution, originality, and the exploitation of works generated by AI take center stage. Researchers are particularly exploring the boundary between human creation and that which is assisted or entirely produced by algorithms, as well as the implications for cultural industries such as music, visual arts, or writing. The distinction between generative and assisted use of AI raises questions about the protection of the resulting works, while the use of protected data to train these systems raises the issue of legal exceptions, such as text and data mining or fair use, depending on the jurisdictions. This concept refers to the possibility of extracting and analyzing data on a large scale for specific uses, under certain legal conditions.
A second axis addresses digital rights, which encompass the management of personal data, the protection of privacy, and rights related to identity. AI systems, capable of reproducing or simulating human traits such as voice or appearance, raise questions about the protection of personality and the commercialization of these attributes. The debates also focus on the impact of AI technologies on freedom of expression, the diversity of online content, and the fight against disinformation, particularly through moderation algorithms or generative models. Ghostbots, systems designed to imitate deceased or absent individuals, illustrate the ethical and legal challenges related to the posthumous use of personal data.
Finally, patent law constitutes the third pillar of this research. Discussions focus on the patentability of inventions generated by AI, the recognition of AI as a potential inventor, and the risks of saturating the patent system through the massive production of prior art. Legal experts also analyze the use of AI as a tool to facilitate the processes of filing, classifying, or evaluating patents, while raising concerns about transparency and accountability in cases of infringement.
The United States and the United Kingdom largely dominate this field of research, with a strong concentration of international collaborations between these countries and a few European or Asian partners. In contrast, contributions from Global South countries remain marginal in indexed publications, reflecting structural inequalities in the production and visibility of academic knowledge. This disparity does not necessarily mean a lack of relevant work in these regions, but rather unequal access to international dissemination platforms.
Despite the adoption of regulatory frameworks, such as European copyright directives or American guidelines, no clear legal consensus has yet emerged. Recent judicial decisions, particularly around the DABUS system, have generally refused to recognize AI as an inventor, reaffirming the need for human involvement for protection under patent or copyright law. Debates on the lawful use of protected data for training AI models remain particularly lively, with divided opinions on the application of fair use or similar exceptions.
The tensions between technological innovation and the protection of existing rights continue to shape discussions at both academic and political levels. The issues also include the remuneration of creators whose works are used to train models, as well as liability in cases of rights violations by generative systems. Ongoing litigation, particularly in the United States, could gradually clarify the legal contours of these practices, but regulatory responses remain fragmented and often limited to non-binding recommendations.
Artificial Intelligence Redefines the Boundaries of Intellectual Property
The academic research on the links between artificial intelligence and intellectual property law has been growing rapidly since 2019, marked by the emergence of generative models. This momentum reflects a global awareness of the legal challenges raised by these technologies, which are profoundly transforming creative and industrial practices.
Three major axes of reflection currently dominate the debates. The first concerns copyright law, where issues of attribution, originality, and the exploitation of works generated by AI take center stage. Researchers are particularly exploring the boundary between human creation and that which is assisted or entirely produced by algorithms, as well as the implications for cultural industries such as music, visual arts, or writing. The distinction between generative and assisted use of AI raises questions about the protection of the resulting works, while the use of protected data to train these systems raises the issue of legal exceptions, such as text and data mining or fair use, depending on the jurisdictions. This concept refers to the possibility of extracting and analyzing data on a large scale for specific uses, under certain legal conditions.
A second axis addresses digital rights, which encompass the management of personal data, the protection of privacy, and rights related to identity. AI systems, capable of reproducing or simulating human traits such as voice or appearance, raise questions about the protection of personality and the commercialization of these attributes. The debates also focus on the impact of AI technologies on freedom of expression, the diversity of online content, and the fight against disinformation, particularly through moderation algorithms or generative models. Ghostbots, systems designed to imitate deceased or absent individuals, illustrate the ethical and legal challenges related to the posthumous use of personal data.
Finally, patent law constitutes the third pillar of this research. Discussions focus on the patentability of inventions generated by AI, the recognition of AI as a potential inventor, and the risks of saturating the patent system through the massive production of prior art. Legal experts also analyze the use of AI as a tool to facilitate the processes of filing, classifying, or evaluating patents, while raising concerns about transparency and accountability in cases of infringement.
The United States and the United Kingdom largely dominate this field of research, with a strong concentration of international collaborations between these countries and a few European or Asian partners. In contrast, contributions from Global South countries remain marginal in indexed publications, reflecting structural inequalities in the production and visibility of academic knowledge. This disparity does not necessarily mean a lack of relevant work in these regions, but rather unequal access to international dissemination platforms.
Despite the adoption of regulatory frameworks, such as European copyright directives or American guidelines, no clear legal consensus has yet emerged. Recent judicial decisions, particularly around the DABUS system, have generally refused to recognize AI as an inventor, reaffirming the need for human involvement for protection under patent or copyright law. Debates on the lawful use of protected data for training AI models remain particularly lively, with divided opinions on the application of fair use or similar exceptions.
The tensions between technological innovation and the protection of existing rights continue to shape discussions at both academic and political levels. The issues also include the remuneration of creators whose works are used to train models, as well as liability in cases of rights violations by generative systems. Ongoing litigation, particularly in the United States, could gradually clarify the legal contours of these practices, but regulatory responses remain fragmented and often limited to non-binding recommendations.
Sources and Credits
Source Study
DOI: https://doi.org/10.1007/s43545-026-01479-5
Title: The state of AI and intellectual property – a thematic scientometric assessment
Journal: SN Social Sciences
Publisher: Springer Science and Business Media LLC
Authors: Gergely Ferenc Lendvai; Peter Mezei; Anett Pogacsas