Our paper “Scenario to specification: promises and pitfalls of AI in developing user-centered engineering specifications with interdisciplinary teams”, written with Prof. KwanMyung Kim, received the Reviewers' Choice Award at ICED25, the International Conference on Engineering Design, in Texas.
The paper compares two interdisciplinary teams working through the Action Steps framework, one with generative-AI support and one without. AI shortened some steps but led to fewer outputs: 8 engineering specifications against 17 in the team without AI. Information overload and a weak grasp of context limited its benefit; it was most useful as a technical consultant and as a bridge between disciplines.
