PROMPT, COMPARE, REFLECT, REPEAT : learning responsible AI use for product design

Tools

Brisco, Ross and Gunn, Kieran and Holliman, Alexander "Freddie"; Barrie, Jeff and Morgan, Thea and Ross, Joel and Selwyn, Becky and McMahon, Chris and Buck, Lyndon and Brisco, Ross and Bohemia, Erik, eds. (2026) PROMPT, COMPARE, REFLECT, REPEAT : learning responsible AI use for product design. In: Exploring Diversity in Engineering & Product Design Education. Proceedings of the International Conference on Engineering and Product Design Education . The Design Society, GBR, pp. 179-184. ISBN 9781912254248 (https://doi.org/10.35199/EPDE.2026.31)

[thumbnail of Brisco-etal-EPDE-2026-learning-responsible-AI-use-for-product-design]
Preview
 Text. Filename: Brisco-etal-EPDE-2026-learning-responsible-AI-use-for-product-design.pdf
Final Published Version
License: Creative Commons Attribution-NonCommercial 4.0 logo
Download (1MB)| Preview

Abstract

Artificial intelligence is rapidly reshaping product design practice and education, creating both opportunities and risks for how future designers are trained. While AI tools can support ideation, documentation and decision-making, there is growing concern that uncritical use may weaken foundational design skills and reflective practice. This paper presents the development and initial evaluation of a participatory, workshop-based educational intervention centred on an AI-powered product design assistant tool. The approach aims to teach responsible and effective AI use by making students active participants in prompt construction, output evaluation and iterative refinement. The tool generates structured early-stage design documentation from user prompts, encouraging consideration of components, methodology, constraints and regulations. An exploratory workshop was conducted with Engineering Design students, combining instruction on prompt crafting with guided experimentation and reflection activities. Feedback indicates that participants found the tool useful for early-stage concept development and for understanding how input structure affects AI output. Features that enabled parameter and prompt manipulation were particularly valued, supporting critical comparison and reflective learning. However, students also highlighted limitations in technical depth, accuracy and verifiability, reinforcing the need for validation and careful task alignment. The findings suggest that participatory AI-building and prompt-design workshops can support AI literacy, critical judgement and responsible adoption in design education, positioning AI as a scaffold for learning rather than a substitute for design thinking.

ORCID iDs

Brisco, Ross ORCID logoORCID: https://orcid.org/0000-0002-8216-9218, Gunn, Kieran and Holliman, Alexander "Freddie" ORCID logoORCID: https://orcid.org/0000-0001-9511-8121; Barrie, Jeff, Morgan, Thea, Ross, Joel, Selwyn, Becky, McMahon, Chris, Buck, Lyndon, Brisco, Ross and Bohemia, Erik
CORE (COnnecting REpositories)