Writing the Future: The Impact of Artificial Intelligence and Knowledge Graphs on the Music Industry

This article investigates the synergistic potential of integrating knowledge graphs and state-of-the-art artificial intelligence models for the fundamental transformation of musical creativity processes. Against the backdrop of the growing generative music market, a critical lack of deep structural control and music-theoretical grounding in existing AI systems becomes apparent, inevitably leading to the production of compositions with limited long-term coherence. The objective of the study is to conceptualize and justify a hybrid model that unites the deterministic precision of knowledge graph structures with the timbral and textural richness of neural-network generative approaches. The methodological foundation comprised the analysis and synthesis of contemporary music-generation techniques, including Transformer architectures (Music Transformer, Museformer), diffusion models (GETMusic), text-to-music systems (MusicGen), as well as empirical experience of applying graph structures within the Teragraph hackathon. The results demonstrate that the hybrid approach effectively overcomes the limitations of both symbolic and exclusively neural methods. In the architecture the knowledge graph establishes a high-level compositional framework which the AI model realizes in the form of a finished audio artifact, providing an unprecedented level of control over the generation process. The conclusion of the study is that the further development of creative music technologies will proceed along the path of human–machine co-creation based on hybrid systems of this nature. The work will be of interest to researchers in the field of computer music, developers of AI systems, as well as composers and producers seeking to master new creative tools.