Exploring Challenges and Opportunities to Support Designers in Learning to Co-create with AI-based Manufacturing Design Tools

Frederic Gmeiner,Humphrey Yang,Lining Yao,Kenneth Holstein,Nikolas Martelaro

DOI: https://doi.org/10.1145/3544548.3580999

2023-03-01

Abstract:AI-based design tools are proliferating in professional software to assist engineering and industrial designers in complex manufacturing and design tasks. These tools take on more agentic roles than traditional computer-aided design tools and are often portrayed as "co-creators." Yet, working effectively with such systems requires different skills than working with complex CAD tools alone. To date, we know little about how engineering designers learn to work with AI-based design tools. In this study, we observed trained designers as they learned to work with two AI-based tools on a realistic design task. We find that designers face many challenges in learning to effectively co-create with current systems, including challenges in understanding and adjusting AI outputs and in communicating their design goals. Based on our findings, we highlight several design opportunities to better support designer-AI co-creation.

Human-Computer Interaction,Artificial Intelligence

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper primarily explores the challenges and opportunities faced by designers when learning to co-create with AI-based design tools. Specifically:

1. **Research Background**:
- Current AI design tools are increasingly used in professional software to assist engineers and industrial designers in completing complex manufacturing and design tasks. These tools are more autonomous than traditional computer-aided design (CAD) tools and are often referred to as "co-creators."
- However, effectively collaborating with these systems requires skills different from those needed for traditional complex CAD tools.

2. **Research Objectives**:
- Explore the specific challenges designers encounter when learning to work collaboratively with AI design tools.
- Analyze how designers overcome these challenges.
- Propose effective strategies to support designers in better collaborating with AI design tools.

3. **Research Methods**:
- Through a series of "think-aloud" studies, observe how trained engineering and architectural designers learn to collaborate with two different AI design tools to complete complex manufacturing design tasks.
- Based on their interactions with the systems and retrospective interviews, it was found that designers generally value AI assistance but face difficulties in effective collaboration and interpreting design outputs.
- Explore how to support designers in better collaborating with AI tools through human-human collaboration studies, including strategies such as providing step-by-step guidance and facilitating design reflection.

4. **Contributions**:
- Provide a set of observed challenges faced by engineering and architectural designers in learning to collaborate with AI in the context of advanced manufacturing design.
- Gain deep insights into designers' needs and expectations for human-machine collaboration tools.
- Highlight design opportunities to better support designers in learning to collaborate with AI tools.