Generative AI prompts for footwear design: Japanese style boosts creativity, sharp features impact feasibility

Why It Matters

This research highlights the critical role of prompt design in leveraging generative AI for creative ideation in product design. Understanding how prompt elements influence output allows designers to more effectively utilize these tools for concept generation and innovation.

Key Finding

The study found that while generative AI can produce footwear concepts, the creativity and feasibility of these concepts are heavily influenced by the prompts used. Explicit stylistic cues, like 'Japanese style,' enhance creativity, whereas overly sharp design elements can reduce practical feasibility. Omitting specific subcategories in prompts also led to less feasible and less well-implemented designs.

Key Findings

• Prompt implementation correlated weakly positively with design creativity and feasibility.
• Prompts in the 'Japanese style' demonstrated significantly higher creativity due to explicit style descriptions.
• Concepts with sharp generative design features exhibited higher creativity but lower feasibility.
• Concepts generated without subcategories showed lower feasibility and prompt implementation.

Research Evidence

Aim: How do different prompt engineering strategies for generative AI tools affect the creativity and feasibility of footwear design concepts?

Method: Experimental study with expert evaluation

Procedure: Seventeen text-to-image prompts were used to generate footwear concepts via Midjourney. Ten distinct generated concepts were then evaluated by seven footwear design experts for creativity and feasibility. Prompt characteristics, including style descriptors and feature complexity, were analyzed against the expert evaluations.

Sample Size: 7 expert evaluators

Context: Footwear design concept generation using generative AI

Design Principle

Prompt engineering is a critical skill for effectively leveraging generative AI in design ideation, requiring a balance between creative exploration and practical constraints.

How to Apply

When using text-to-image AI for design concept generation, explicitly define desired styles, aesthetics, and functional considerations within the prompts. Iterate on prompts to balance novelty with manufacturability.

Limitations

The study focused on a single AI tool (Midjourney) and a specific product category (footwear), potentially limiting generalizability. Expert evaluations are subjective, and the correlation between prompt implementation and outcomes was weak.

Student Guide (IB Design Technology)

Simple Explanation: Using AI to design shoes works best when you tell the AI exactly what style you want (like 'Japanese style') and be careful not to ask for designs that are too complex or pointy, as those might be hard to actually make.

Why This Matters: This research shows how to use AI tools more effectively in your design projects by understanding how to write better instructions (prompts) to get the results you want, especially for generating new ideas.

Critical Thinking: To what extent can AI truly be 'creative,' or is it merely a sophisticated tool for recombining existing data based on user input?

IA-Ready Paragraph: The effectiveness of generative AI in design ideation is significantly influenced by prompt engineering. Research by Cheng (2023) indicates that incorporating specific stylistic descriptors, such as 'Japanese style,' can enhance the creativity of AI-generated footwear concepts, while overly complex or sharp design features may decrease their feasibility. This suggests that designers must carefully craft prompts to balance novelty with practical considerations when using AI for concept development.

Project Tips

• When using AI for concept generation, document your prompts meticulously.
• Consider how you will evaluate the 'creativity' and 'feasibility' of AI-generated outputs in your design project.

How to Use in IA

• Reference this study when discussing the use of AI tools for ideation and concept generation in your design project, particularly when analyzing the effectiveness of different prompt strategies.

Examiner Tips

• Ensure your analysis of AI-generated concepts clearly links prompt characteristics to observed outcomes, rather than just presenting raw outputs.

Independent Variable: ["Prompt characteristics (e.g., style descriptors, feature complexity, inclusion of subcategories)"]

Dependent Variable: ["Design creativity","Design feasibility","Prompt implementation"]

Controlled Variables: ["AI tool used (Midjourney)","Product category (footwear)","Number of generated concepts"]

Strengths

• Employs expert evaluation for assessing design quality.
• Investigates specific prompt engineering techniques.
• Focuses on a relevant emerging technology in design.

Critical Questions

• How might the training data of the AI model influence the 'creativity' and 'feasibility' of its outputs?
• What are the ethical implications of relying on AI for design ideation, particularly concerning originality and authorship?

Extended Essay Application

• An Extended Essay could explore the impact of different AI models on design ideation, or investigate the optimal prompt structures for specific design disciplines beyond footwear.

Source

Impact of Generative Artificial Intelligence on Footwear Design Concept and Ideation · 2023 · 10.3390/engproc2023055032