The Art That Poses Back: Assessing AI Pastiches after Contemporary Artworks
arXiv:2603.06324v1 Announce Type: new Abstract: This study explores artificial visual creativity, focusing on ChatGPT's ability to generate new images intentionally pastiching original artworks such as paintings, drawings, sculptures and installations. The process involved twelve artists from Romania, Bulgaria, France, Austria, and the United Kingdom, each invited to contribute with three of their artworks and to grade and comment on the AI-generated versions. The analysis combines human evaluation with computational methods aimed at detecting visual and stylistic similarities or divergences between the original works and their AI-produced renditions. The results point to a significant gap between color and texture-based similarity and compositional, conceptual, and perceptual one. Consequently, we advocate for the use of a "style transfer dashboard" of complementary metrics to evaluate the similarity between pastiches and originals, rather than using a single style metric. The artist
arXiv:2603.06324v1 Announce Type: new Abstract: This study explores artificial visual creativity, focusing on ChatGPT's ability to generate new images intentionally pastiching original artworks such as paintings, drawings, sculptures and installations. The process involved twelve artists from Romania, Bulgaria, France, Austria, and the United Kingdom, each invited to contribute with three of their artworks and to grade and comment on the AI-generated versions. The analysis combines human evaluation with computational methods aimed at detecting visual and stylistic similarities or divergences between the original works and their AI-produced renditions. The results point to a significant gap between color and texture-based similarity and compositional, conceptual, and perceptual one. Consequently, we advocate for the use of a "style transfer dashboard" of complementary metrics to evaluate the similarity between pastiches and originals, rather than using a single style metric. The artists' comments revealed limitations of ChatGPT's pastiches after contemporary artworks, which were perceived by the authors of the originals as lacking dimensionality, context, and intentional sense, and seeming more of a paraphrase or an approximate quotation rather than as a valuable, emotion-evoking artwork.
Executive Summary
This study explores the ability of ChatGPT to generate new images intentionally pastiching original artworks. Twelve artists contributed their works, grading and commenting on AI-generated versions. The analysis reveals a gap between color and texture-based similarity and compositional, conceptual, and perceptual similarity. The findings suggest the use of a 'style transfer dashboard' to evaluate similarity between pastiches and originals. However, the artists' comments highlight limitations in ChatGPT's pastiches, perceiving them as lacking dimensionality, context, and intentional sense. The study underscores the need for a more comprehensive approach to evaluating AI-generated art, combining human evaluation and computational methods. The results have practical implications for the art world, where AI-generated art is increasingly prevalent, and policy implications for the fair use and ownership of AI-generated works.
Key Points
- ▸ ChatGPT's ability to generate pastiches of original artworks is explored in the study.
- ▸ Human evaluation and computational methods are combined to assess similarity between pastiches and originals.
- ▸ The results highlight limitations in ChatGPT's pastiches, particularly in terms of dimensionality, context, and intentional sense.
Merits
Strength in Methodology
The study combines human evaluation and computational methods, providing a comprehensive approach to assessing AI-generated art.
Insight into AI-generated Art
The study sheds light on the limitations of ChatGPT's pastiches, highlighting the need for a more nuanced understanding of AI-generated art.
Demerits
Limited Sample Size
The study involves a relatively small sample size of 12 artists, which may limit the generalizability of the findings.
Dependence on ChatGPT
The study relies on ChatGPT's ability to generate pastiches, which may not be representative of other AI algorithms or techniques.
Expert Commentary
The study's findings have significant implications for the art world, where AI-generated art is increasingly prevalent. The study's focus on the limitations of ChatGPT's pastiches highlights the need for a more nuanced understanding of AI-generated art. The combination of human evaluation and computational methods provides a comprehensive approach to assessing AI-generated art. However, the study's limitations, including the small sample size and dependence on ChatGPT, should be considered in future research. Overall, the study makes a valuable contribution to the growing body of research on AI-generated art, and its findings have important implications for both practical and policy considerations.
Recommendations
- ✓ Future research should focus on replicating the study with a larger sample size and incorporating other AI algorithms and techniques.
- ✓ Artists, curators, and critics should be aware of the limitations of AI-generated art and consider additional contextual information when evaluating its meaning and intent.