인공지능 활용 디지털 치료제의 지식재산권 문제 - 인공지능 생성 음악의 저작권 문제를 중심으로 -

Digital therapeutics (DTx) represent a transformative intersection between traditional medical interventions and digital technologies, offering substantial market potential and growth opportunities across diverse sectors, including music therapy. As music therapy increasingly gains empirical validation for its efficacy in addressing physical and mental health conditions, its integration with DTx is becoming more prominent. The rapid advancements in artificial intelligence (AI), particularly generative AI, further underscore the transformative potential of AI-driven DTx in content generation, including therapeutic music. Intellectual property (IP) law has historically been instrumental in fostering innovation by providing robust frameworks for protecting creative outputs. This study critically examines the IP implications associated with AI-powered DTx, with a particular emphasis on copyright issues. The intersection of AI and copyright law has been the subject of extensive scholarly debate, particularly concerning potential copyright infringement arising from the reproduction of copyrighted works during AI model training. Various solutions, including legislative interventions, have been proposed to address these challenges. Notably, jurisdictions such as Japan and the European Union have introduced text and data mining (TDM) exceptions, which have sparked significant discourse regarding their practical implications and limitations. In parallel to debates on the permissibility of using copyrighted works for AI model training, there has been growing advocacy for recognizing the economic and creative value of AI-generated outputs. As IP frameworks are designed to protect valuable innovations, extending protection to AI-generated works invites further exploration. These debates are particularly relevant to AI-based DTx, where the generation of therapeutically effective music may necessitate the use of pre-existing copyrighted materials. Simultaneously, the outputs of generative AI systems tailored for therapeutic purposes could themselves be candidates for IP protection. However, the distinctive characteristics of DTx—primarily their goal-oriented focus on therapeutic efficacy—may reduce reliance on pre-existing musical works. Moreover, the market for music-based DTx is expected to be dominated by a limited number of specialized players, leading to further consolidation and niche development. While the creation of a bespoke IP framework specifically for music-based DTx may be impractical, it is critical to explore appropriate protection mechanisms and evaluate various legal strategies. As these technologies continue to evolve, it is essential to balance the dual imperatives of fostering innovation in DTx and addressing complex IP challenges, ensuring their meaningful contribution to advancing healthcare.