E-Thesis 17 views 6 downloads
Co-Creativity at the Intersection of Place, Poetry, and Generative AI / IBUKUN OLATUNJI
Swansea University Author: IBUKUN OLATUNJI
-
PDF | E-Thesis – open access
Copyright: the author, Ìbùkún O látúnjí, 2026. - Distributed under the terms of a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).
Download (57.08MB)
DOI (Published version): 10.23889/SUThesis.71618
Abstract
This thesis investigates how generative AI can support and reshape human creativity across place, poetry, and performance. It makes three original contributions: (i) introducing cultural geography as a framework for designing and evaluating AI-enabled creative infrastructure; (ii) developing computa...
| Published: |
Swansea
2026
|
|---|---|
| Institution: | Swansea University |
| Degree level: | Doctoral |
| Degree name: | Ph.D |
| Supervisor: | Jones, M., Rahat, A., and Rogers, A. |
| URI: | https://cronfa.swan.ac.uk/Record/cronfa71618 |
| Abstract: |
This thesis investigates how generative AI can support and reshape human creativity across place, poetry, and performance. It makes three original contributions: (i) introducing cultural geography as a framework for designing and evaluating AI-enabled creative infrastructure; (ii) developing computational proxies such as rhyme density and cadence similarity for assessing poetic and performance creativity; and (iii) pioneering adversarial human–AI interaction as a method for evaluating co-creativity. Methodologically, it integrates spatial inquiry, poetic analysis, and design-based evaluation, positioning creativity as a relational practice that emerges across human, machine, and cultural contexts. Grounded in hip hop principles, Study I examines how physical environments shape creativity through the design and activation of cultural venues in Swansea and London. The study demonstrates how purpose-built infrastructure can function as a test bed for digital cultural systems, a perspective rarely addressed in AI research. Study II investigates AI-assisted poetic composition using computational writing tools. Structured tasks, participant evaluations, and rhyme density as a novel proxy are employed to assess how system types influence creativity. Study III extends into spoken-word performance, presenting a systematic evaluation of AI in freestyle rap. Six experiments assess voice clone similarity, cadence accuracy, and text generation, combining computational analysis with human judgements. The findings demonstrate cadence as a critical dimension for human–AI co-performance. Together, the three studies contribute to a layered model of human–AI co-creativity, spanning spatial, textual, and performance domains. The thesis positions adversarial evaluation as its central contribution, demonstrating how creative sparring provokes originality. Stylistic and improvisational dimensions serve as complementary modes of analysis, offering interpretable lenses on the co-creative process. |
|---|---|
| Keywords: |
Human–AI Co-Creativity, Computational CreativityGenerative AI Evaluation, Expressive Language Modelling, Rhyme Density Metrics, Cadence Similarity. Digital Cultural Infrastructure, Creative Placemaking, Digital Creative Twins |
| College: |
Faculty of Science and Engineering |
| Funders: |
EPSRC, Swansea Council |