Beyond the Music Video: AI, Worldbuilding, and Audiovisual Music Practice
Hermes, K. 2026. Beyond the Music Video: AI, Worldbuilding, and Audiovisual Music Practice. Innovation in Music 26. Aalborg University, Denmark 12 - 14 Jun 2026 Routledge.
Hermes, K. 2026. Beyond the Music Video: AI, Worldbuilding, and Audiovisual Music Practice. Innovation in Music 26. Aalborg University, Denmark 12 - 14 Jun 2026 Routledge.
| Title | Beyond the Music Video: AI, Worldbuilding, and Audiovisual Music Practice |
|---|---|
| Authors | Hermes, K. |
| Abstract | Musicians increasingly operate within audiovisual ecosystems, where the production of visual materials is integral to musical practice. While music videos have historically functioned as discrete promotional artefacts tied to a single release, contemporary music practices increasingly demand continuous and modular forms of audiovisual production. Musicians now operate within platform-driven ecosystems such as Instagram, TikTok, YouTube and live-streaming environments that prioritise frequent updates, short-form visuals, and ongoing audience engagement over singular, canonical works (Vernallis, 2004; Korsgaard, 2017). This shift places additional pressure on musicians, who need to maintain a coherent audiovisual presence alongside musical production, often without access to specialised visual teams (Hermes, 2025). This paper explores how recent developments in AI-supported tools, 3D environments, and generative workflows enable musicians to move beyond the bounded format of the music video towards ongoing processes of audiovisual world-building. The author reflects on emerging audiovisual workflows that combine music production with 3D tools and AI-assisted asset generation, particularly for contexts where visual environments are actively designed rather than simply recorded. These workflows enable musicians to develop stories, personae, and environments that can evolve across releases, performances, and platforms. Such approaches blur distinctions between composition, visual design, performance, and marketing. This approach foregrounds questions of authorship, control, and creative coherence, particularly in relation to prompt-based systems and machine learning models that introduce stochastic or opaque processes into creative decision-making. Rather than positioning AI as a replacement for human creativity, the paper adopts a human-centred perspective (Shneiderman, 2022), examining how AI can be integrated as assistive, generative, and exploratory tools within musicians’ existing practices. In this context, AI is framed not as an autonomous creative agent but as part of a heterogeneous creative infrastructure that supports ideation, variation, and iteration. Drawing on several case studies, the paper adopts an exploratory stance, identifying design tensions and opportunities that arise when musicians engage with AI-supported worldbuilding. These include issues of speed versus depth, automation versus agency, and consistency versus variation within audiovisual identities. The paper argues for the value of flexible, modular workflows that leave space for experimentation and future technical development, including more advanced machine learning approaches. Ultimately, this paper positions audiovisual world-building as an emerging mode of musical practice, one that extends beyond the music video while remaining grounded in musicians’ creative intentions, aesthetic judgement, and lived experience. |
| Keywords | Worldbuilding; Generative AI; Computational creativity; Co-creativity; Practice-led case studies; AI-powered music video |
| Year | 2026 |
| Conference | Innovation in Music 26 |
| Publisher | Routledge |
| Publisher's version | File Access Level Open (open metadata and files) |