Stable diffusion models reveal a persisting human and AI gap in visual creativity

2025-11-20

AI Summary

This study investigates the performance gap in visual creativity between humans and generative AI (Stable Diffusion), and how human guidance modulates AI output quality. Method: We conducted a controlled experiment with three human groups—professional artists, non-artist adults—and two AI prompting conditions (“human-inspired” vs. “self-guided”)—evaluated by 255 human raters and GPT-4o across multiple creativity dimensions. Contribution/Results: Visual creativity exhibits a significant hierarchical gradient: artists > non-artists > human-inspired AI > self-guided AI. Human guidance substantially enhances AI-generated outputs, elevating them to near non-artist human levels. Critically, this work provides the first empirical evidence that human perceptual granularity and contextual sensitivity remain irreplaceable in visual ideation, while exposing fundamental limitations in large language models’ cross-modal transfer capability—particularly in grounding linguistic prompts in visual semantics. These findings establish both theoretical foundations and practical frameworks for human-AI co-creative systems.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

While recent research suggests Large Language Models match human creative performance in divergent thinking tasks, visual creativity remains underexplored. This study compared image generation in human participants (Visual Artists and Non Artists) and using an image generation AI model (two prompting conditions with varying human input: high for Human Inspired, low for Self Guided). Human raters (N=255) and GPT4o evaluated the creativity of the resulting images. We found a clear creativity gradient, with Visual Artists being the most creative, followed by Non Artists, then Human Inspired generative AI, and finally Self Guided generative AI. Increased human guidance strongly improved GenAI's creative output, bringing its productions close to those of Non Artists. Notably, human and AI raters also showed vastly different creativity judgment patterns. These results suggest that, in contrast to language centered tasks, GenAI models may face unique challenges in visual domains, where creativity depends on perceptual nuance and contextual sensitivity, distinctly human capacities that may not be readily transferable from language models.

Problem

• AI struggles with visual creativity compared to humans
• Human guidance improves AI's visual creative output
• Human and AI raters differ in creativity judgment patterns

Innovation

• Used human-inspired and self-guided AI prompting techniques
• Employed human and GPT4o raters for creativity evaluation
• Revealed creativity gap between human artists and AI models