Computational Power and Subjective Quality of AI-Generated Outputs: The Case of Aesthetic Judgement and Positive Emotions in AI-Generated Art

Simone Grassini
DOI: https://doi.org/10.1080/10447318.2024.2422755
IF: 4.92
2024-11-13
International Journal of Human-Computer Interaction
Abstract:As AI technology advances, important questions arise about how the computational power behind these models affects human perception of AI-generated contents. Furthermore, the significant energy consumption of resource-heavy AI models brings growing environmental concerns as the demand for computational resources continues to rise. This research examines the influence of computational power in AI-generated art on human aesthetic perception and emotional response. A group of 100 participants evaluated several artworks either produced by humans or created by an AI operating at three distinct levels of quality, corresponding to varying computational demands. The results showed that participants preferred AI-generated artworks created with greater computational effort, indicating higher likability and more positive emotional reactions. However, the improvements in perception and emotion were modest and exhibited diminishing returns as computational power increased. The study addresses ethical concerns, including the environmental impact of the large computational resources generative AI requires and its implications for sustainable AI development.
computer science, cybernetics,ergonomics
What problem does this paper attempt to address?