Dynamics of collective creativity in AI art competitions
The study explores the dynamics of collective creativity in AI art competitions, specifically through the platform Artbreeder. It analyzes how users remix images over time, revealing trends in simplicity and thematic convergence. The findings indicate that while novel images lead to more complex offspring, users tend to prefer less novel works for remixing.
- ▪The research analyzed a dataset of 130,882 images from 368 remix parties over 13 months.
- ▪Images produced in these competitions tend to become simpler and converge toward common themes.
- ▪Larger remix parties generate more novelty but at the expense of complexity.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.17141 (cs) [Submitted on 16 May 2026] Title:Dynamics of collective creativity in AI art competitions Authors:Mason Youngblood, Jeff Nusz, Joel Simon View a PDF of the paper titled Dynamics of collective creativity in AI art competitions, by Mason Youngblood and 2 other authors View PDF Abstract:Creativity is a fundamental aspect of how culture evolves, yet the mechanisms by which groups produce novelty are notoriously difficult to infer from the historical record.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
