Generative AI's Reconstruction of Music Creation Practices Through Media Affordances: A Case Study of User Interaction on Suno

Abstract

Generative AI technologies are developing very fast these days. AI music platforms like Suno are slowly changing the traditional ways of music creation. But most existing studies mainly focus on technical abilities, copyright disputes and influences on the music industry, while few of them pay attention to how generative AI reshapes users' creative activities through media platform structures. For this reason, this paper takes Suno as its research object. With the help of media affordance theory, it uses methods of case analysis and participatory observation, together with practical experiments of writing prompts, to study how generative AI platforms rebuild music creation practices during user interaction. The study finds that Suno forms a new logic of music creation through four types of media affordance: functional availability, rule visibility, system persistence and relational variability. First, functional availability changes the foundation of music creation. It shifts from reliance on music theory skills to dependence on control through words. Second, the platform's rules are not fully clear to its users. This leads users to develop new creative strategies by trying different working rules of the system. Third, system persistence changes the whole creative process. It turns the process from a one-off linear task to repeated generation and improvement. Fourth, relational variability promotes the change of creators' identity. Creators are no longer independent individual subjects, but work in the mode of human-AI cooperation. This study holds that generative AI has not taken the place of music creators. Instead, through media affordances, it has changed the way music creation works. It makes music production a more and more dynamic practice, in which people keep generating works and making choices during human-machine interaction.