"We Are Visual Thinkers, Not Verbal Thinkers!": A Thematic Analysis of How Professional Designers Use Generative AI Image Generation Tools

Authors. Hyerim Park, Joscha Eirich, Andre Luckow, Michael Sedlmair

Venue. NordCHI (2024) Full Paper

Type. Full Paper

Materials. DOI ACM PDF

Abstract. Generative artificial intelligence (GenAI) has become increasingly popular, influencing various creative domains. However, while broader societal perspectives have been analyzed, specific examinations of how practitioners utilize GenAI tools to enhance their current workflows remain limited. To address this gap, we conducted a qualitative study involving 16 professional designers from the automotive industry. We aimed to identify their challenges with existing GenAI image generation tools in daily design practices. Thematic analysis revealed four key themes: (1) the need for visual input-centric multi-modal interfaces that extend beyond textual prompts, (2) the lack of support for the iterative nature of design processes in GenAI tools, (3) difficulties in controlling prompts to achieve desired outputs, and (4) the significance of incorporating human experiences and emotions into design. Based on our findings, we propose and discuss potential design considerations for enhancing future GenAI image generation tool interfaces.

@inproceedings{10.1145/3679318.3685370,
    title        = {``We Are Visual Thinkers, Not Verbal Thinkers!'': A Thematic Analysis of How Professional Designers Use Generative {AI} Image Generation Tools},
    author       = {Park, Hyerim and Eirich, Joscha and Luckow, Andre and Sedlmair, Michael},
    year         = {2024},
    booktitle    = {Proceedings of the 13th Nordic Conference on Human-Computer Interaction},
    publisher    = {Association for Computing Machinery},
    series       = {NordiCHI '24},
    doi          = {10.1145/3679318.3685370},
    url          = {https://doi.org/10.1145/3679318.3685370},
    abstract     = {Generative artificial intelligence (GenAI) has become increasingly popular, influencing various creative domains. However, while broader societal perspectives have been analyzed, specific examinations of how practitioners utilize GenAI tools to enhance their current workflows remain limited. To address this gap, we conducted a qualitative study involving 16 professional designers from the automotive industry. We aimed to identify their challenges with existing GenAI image generation tools in daily design practices. Thematic analysis revealed four key themes: (1) the need for visual input-centric multi-modal interfaces that extend beyond textual prompts, (2) the lack of support for the iterative nature of design processes in GenAI tools, (3) difficulties in controlling prompts to achieve desired outputs, and (4) the significance of incorporating human experiences and emotions into design. Based on our findings, we propose and discuss potential design considerations for enhancing future GenAI image generation tool interfaces.},
    articleno    = {35},
    numpages     = {14},
    keywords     = {creativity support tools, generative AI, human-AI interaction, qualitative research}
}