Using machine learning to learn machines: a cross-cultural study of users’ responses to machine-generated artworks

Abstract

Drawing upon prior literature on machine-generated news, this study expands the research scope to machine-generated art works in a cross-cultural context. This study combines machine learning approaches with online experiments and investigates how different genres of art works and different authorship cues influence participants’ open-ended responses to machine-generated works. Results suggest that while genres and cultures affected participants’ discussion topics and word use, the differences between participants’ responses to machine-generated art works and human-generated ones were not evident. This study tests the explanatory power of machine heuristic and demonstrates the feasibility of integrating multiple methods in future AI-based media research.