This study investigates the capabilities and limitations of artificial intelligence (AI) image generation platforms in simulating the stylistic features of traditional Chinese Gongbi painting. Focusing on four widely used platforms—leonardo.ai, Doubao, ChatGPT, and Artbreeder—the research conducts a comparative analysis across two key modalities: text-to-image and image-to-image generation. Using parameters such as composition, line quality, color treatment, and style fidelity, the study evaluates each platform’s performance in reproducing the refined aesthetics of Gongbi painting. The classic Chinese art theory Six Principles of Xie He is employed as a critical framework, assessing the AI-generated images through six dimensions: spirit resonance (qiyun), structural brushwork (gufa), formal likeness, color application, compositional arrangement, and stylistic imitation. Findings reveal that while AI systems demonstrate competence in replicating visual structures and color harmonies—making them suitable for preliminary sketching and stylistic exploration—they struggle with the nuanced expression of brush rhythm, cultural semantics, and artistic intentionality that define traditional Chinese painting. The study concludes that although current AI platforms cannot replace human artists as primary creative agents, they serve as valuable tools for inspiration, stylistic experimentation, and augmenting traditional workflows. This research thus contributes to a deeper understanding of how AI can interact with and extend the practices of classical art in the contemporary technological context.
M. A. Sahir Ar kan (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: