14 Generative AI-driven Design: A Framework for Enhancing Cross-disciplinary Collaborative Creativity

Abstract

Traditional design processes in cross-disciplinary teams often face challenges in communication, knowledge integration, and early-stage ideation, leading to suboptimal creative outcomes. While AI tools are emerging, there is a lack of a systematic framework to integrate them effectively into collaborative creative workflows. This study proposes a novel framework, "Generative AI-driven Collaborative Creativity" (GA-CC), which leverages Large Language Models (LLMs) to facilitate brainstorming, concept development, and knowledge sharing among designers, engineers, and business strategists. The framework's effectiveness is evaluated through a mixed-methods approach, including a controlled experiment with 60 participants from diverse professional backgrounds, qualitative analysis of collaborative sessions, and quantitative assessment of creative outputs using the consensual assessment technique (CAT). The results demonstrate that teams using the GA-CC framework produced concepts that were rated significantly higher in novelty, usefulness, and feasibility compared to control groups. The framework also improved interdisciplinary communication and accelerated the convergence of ideas. This research provides empirical evidence for the value of integrating generative AI into the conceptual design phase and offers a practical framework for organizations to enhance their innovative capabilities. It contributes to the fields of design science, human-computer interaction, and engineering management by offering a new paradigm for technology-mediated creative collaboration.