Patterns of Creativity: How User Input Shapes AI-Generated Visual Diversity

TL;DR

This research examines how users' tendencies to create original prompts or rely on common templates influence content homogenization during interactions with Text-to-Image models, and developed three originality metrics and applied them to user-generated prompts from two datasets, DiffusionDB and Civiverse.

Abstract

Recent critiques of Artificial-intelligence (AI)-generated visual content highlight concerns about the erosion of artistic originality, as these systems often replicate patterns from their training datasets, leading to significant uniformity and reduced diversity. Our research adopts a novel approach by focusing on user behavior during interactions with Text-to-Image models. Instead of solely analyzing training data patterns, we examine how users' tendencies to create original prompts or rely on common templates influence content homogenization. We developed three originality metrics -- lexical, thematic, and word-sequence originality -- and applied them to user-generated prompts from two datasets, DiffusionDB and Civiverse. Additionally, we explored how characteristics such as topic choice, language originality, and the presence of NSFW content affect image popularity, using a linear regression model to predict user engagement. Our research enhances the discourse on AI's impact on creativity by emphasizing the critical role of user behavior in shaping the diversity of AI-generated visual content.

Figures (11)

• Figure 1: Categorization of the primary topics and their relative proportions within the DiffusionDB dataset.
• Figure 2: Categorization of the primary topics and their relative proportions within the Civiverse dataset.
• Figure 4: UMAP-based visualization of image clusters from CLIP embeddings.
• Figure 5: UMAP-based visualization of text prompt clusters from CLIP embeddings.
• Figure 6: Portrait images generated from prompts low in lexical originality.