Data mining is a technical process of analyzing data for the purpose of extracting information and patterns or generating new works and outputs. The data subject to this process may be protected by intellectual property rights, and their use without authorization may expose data miners to claims of intellectual property infringement. Accordingly, the central question is whether data mining, in itself, constitutes an infringement of intellectual property rights and whether it may fall within the scope of exceptions to such rights. This issue has become particularly significant and complex with the rapid development of artificial intelligence, as advances in artificial intelligence and machine learning are closely associated with access to and large-scale processing of data, a substantial portion of which may be protected by intellectual property rights.

This paper examines data mining in light of the three-step test, the fair use doctrine, and the exceptions provided for under national laws, with a view to determining whether data mining may constitute a normal exploitation of a work and whether it causes unreasonable prejudice to the legitimate interests of the right holder. Employing a descriptive-analytical methodology and emphasizing the need to strike a balance between the protection of intellectual property rights and technological development, the study concludes that a distinction should be drawn between the data-mining process and its resulting outputs. Although certain outputs generated through data mining may infringe copyright in literary and artistic works, the data-mining process itself may, subject to the fulfillment of the conditions established by applicable exceptions and relevant interpretive criteria, be regarded as a permissible use of protected works.