Dynamics of collective creativity in AI art competitions

2026-05-16

AI Summary

This study investigates the mechanisms underlying collective novelty generation and the manifestations of cultural evolution within networked human–AI collaborative systems. Leveraging large-scale user behavior and image data from Artbreeder’s “Remix Party” platform, the research integrates computational image feature quantification, cultural evolutionary modeling, and social computing methods to reveal, for the first time, the dynamic patterns of collective creativity in real-world human–AI artistic co-creation networks. The findings demonstrate that image evolution tends toward simplification and convergence around thematic attractors; while offspring images derived from highly novel parent images exhibit greater complexity and receive more likes, users preferentially remix less novel works. Moreover, although larger parties enhance overall novelty, they simultaneously reduce image complexity, uncovering a paradoxical tension between novelty and user preference.

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. Iterated learning experiments have shown that cultural transmission reliably distorts artifacts toward the inductive biases of learners, but most of this work uses linear chains between human participants, leaving open how these dynamics play out in the networked, human-AI systems that increasingly shape cultural production. In this study, we leverage one such system, Artbreeder, which hosts daily "remix parties" where users iteratively build on each other's work from a single seed image, producing branching lineages of human-AI co-created images. We analyze a dataset of 130,882 images from 368 remix parties over 13 months and find that images become simpler and converge toward common thematic "attractors" (e.g., steampunk scenes, alien architecture). We also find that while more novel "parent" images produce more novel and complex "children" that attract more likes, users paradoxically prefer to remix images that are less novel and complex. Finally, larger remix parties produce more novelty at the cost of lower complexity.

Problem

• collective creativity
• AI art
• cultural transmission
• novelty
• human-AI collaboration

Innovation

• collective creativity
• human-AI collaboration
• iterated learning
• cultural evolution
• computational aesthetics