Facilitating Design Knowledge Access With a Generative AI-Based Conversational Assistant

Exploring the knowledge base to identify relevant existing scientific knowledge is one of the first activities in any Design Science Research (DSR) project. Traditionally, researchers relied on search engines in conducting these activities. A more recent and promising approach to increase the effectiveness and efficiency of search activities is the use of AI. However, despite the promises, there are challenges in using AI assistants in knowledge-intensive tasks, such as fabricating false or misleading responses and offering suggestions with scant explanations. Addressing these problems in the specific context of DSR, we propose a generative AI-based conversational assistant to facilitate exploring existing design knowledge and support knowledge reuse. Following the DSR paradigm, we present the results of the first design cycle, including a set of design requirements, and propose design principles. Additionally, we present a prototypical instantiation of the design principles and report the results of a formative evaluation.