Understanding Design Fixation in Generative AI

2025-02-09

AI Summary

Generative AI (GenAI) suffers from “design fixation”—a phenomenon wherein models, over-reliant on frequent patterns in training data, produce design outputs lacking novelty and diversity—hindering human creativity in design tasks. Method: We propose the first systematic theoretical framework that rigorously defines design fixation’s conceptual scope, root causes, manifestations, and impacts, distinguishing its characteristics across text-based large language models and image-based diffusion models. Combining formal theoretical modeling with multi-task empirical experiments, we uncover cross-modal, task-agnostic mechanisms underlying fixation. Contribution/Results: The study identifies universal drivers of fixation across modalities and tasks, and introduces actionable mitigation strategies—including controllable latent-space diversification and human-in-the-loop feedback integration. These advances establish a novel paradigm for next-generation human-AI co-design tools that actively augment, rather than constrain, creative cognition. Our framework bridges theoretical rigor with practical implementation, offering both foundational insights and deployable guidelines for designers and AI developers.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Generation

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
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Generative AI (GenAI) provides new opportunities for creativity support, but the phenomenon of GenAI design fixation remains underexplored. While human design fixation typically constrains ideas to familiar or existing solutions, our findings reveal that GenAI similarly experience design fixation, limiting its ability to generate novel and diverse design outcomes. To advance understanding of GenAI design fixation, we propose a theoretical framework includes the definition, causes, manifestations, and impacts of GenAI design fixation for creative design. We also conducted an experimental study to investigate the characteristics of GenAI design fixation in practice. We summarize how GenAI design fixation manifests in text generation model and image generation model respectively. Furthermore, we propose methods for mitigating GenAI design fixation for future creativity support tool design. We recommend adopting the lens of GenAI design fixation for creativity-oriented HCI research, as the unique perspectives and insights it provides.

Problem

• Explores GenAI design fixation phenomenon
• Proposes framework for understanding GenAI fixation
• Suggests methods to mitigate design fixation

Innovation

• Theoretical framework for GenAI fixation
• Experimental study on fixation characteristics
• Methods to mitigate GenAI design fixation