上传
生成语法
视觉艺术
编码(社会科学)
计算机科学
万维网
社会学
艺术
人工智能
社会科学
作者
Qingyu Guo,Kangyu Yuan,Changyang He,Zhenhui Peng,Xiaojuan Ma
标识
DOI:10.1145/3613905.3650851
摘要
The rise of AI-generated content (AIGC) is transforming online creative communities (OCCs) and posing challenges to their regulation. Artwork description may reveal creators' practice and motivation in creating and sharing artwork. Understanding the influence of AIGC on creators' descriptions of shared artwork could be helpful for community regulation. In this work, we collect 235K posts from DeviantArt, a large creative community that allows uploading AIGC. We confirm the prevalence of AIGC in the community. Through an open coding on 800 randomly sampled posts, we identify five themes in artwork descriptions. We quantitatively examine how these themes are affected by the prevalence of AIGC via statistical analysis. Results indicate a shift towards commercial opportunities and a reduced focus on copyright since the prevalence of AIGC. Descriptions for AI-generated artworks are more likely to direct members to other creations than those for human-created artworks. Finally, we discuss insights for OCCs.
科研通智能强力驱动
Strongly Powered by AbleSci AI