Stars Interwoven: A Cross-Cultural Study of Machine Imagination and Computational Aesthetics

Source of Publication

Proceedings of the 31st International Symposium on Electronic/Emerging Art (ISEA2026)

Conference Year

2026

Abstract

This study examines the production process of the artwork Stars Interwoven and explores whether generative artificial intelligence can translate cross-cultural textual references to the cosmos into coherent visual experience. Unlike prompt-driven creation workflows, this project develops a curated corpus of “star-related texts” drawn from twelve cultural contexts—poems, prayers, lullabies, folk songs, and mythic fragments in which the star is central. Through a process of AI-assisted selection and multimodal generation, we observe how the model interprets and transforms language carrying cultural memory into visual form, identifying patterns of misreading, recompositing, poetic inference, and latent bias. The work is not only concerned with the generated images themselves, but also treats the act of generation as a research site to examine how machines reorganize mythology, metaphor, and symbolic imagination. Through a small-scale cross-cultural artistic experiment, this project proposes a possible perspective for understanding machine imagination, cultural abstraction, and algorithmic tendencies in contemporary creative practice.

Document Type

Conference Proceeding

Publication Date

11-6-2026

This document is currently not available here.

Share

COinS