<![CDATA[This study aims to identify trends and developments in scientific research on AI-generated graphics during the period 2020–2025 using a bibliometric approach. The analysis focuses on publication trends, institutional and country collaborations, thematic distributions, and research network visualizations. Bibliometric data were collected from the Scopus database, resulting in 4,968 scientific documents retrieved using keywords such as “AI-generated graphics,” “text-to-image,” “DALL-E,” “Stable Diffusion,” and “Midjourney.” The analysis was conducted using VOSviewer to examine keyword co-occurrence, co-authorship, co-citation, and bibliographic coupling. The results show a significant increase in publications, with a notable peak in 2024. China emerged as the leading contributor, followed by the United States, supported by major research funding institutions such as the National Natural Science Foundation of China (NSFC). Thematic analysis indicates a shift from GAN-based models toward diffusion-based text-to-image models, accompanied by growing attention to semantic, ethical, and social issues. Network visualization reveals ten major research clusters, including health, digital arts, and computer vision. Unlike previous bibliometric studies that primarily focus on general artificial intelligence or isolated generative models, this study provides a comprehensive and up-to-date mapping of AI-generated graphics research by integrating technological evolution, creative AI platforms, and ethical dimensions within a single analytical framework. These findings offer a holistic understanding of the research landscape and serve as a foundation for future technological development and interdisciplinary collaboration in AI-generated graphics.]]>
Copyrights © 2025
