The Design, Development, and Evaluation of Artificial Intelligence (AI) Assistant Prototypes of Form Concept Ideation for Architectural Design
Access status:
USyd Access
Type
ThesisThesis type
Doctor of PhilosophyAuthor/s
Dai, ShuyaoAbstract
The rapid advancement of Generative Artificial Intelligence presents a paradigm shift for the architectural discipline by offering capabilities for form generation and visual exploration. However the integration of these technologies into the architectural workflow remains challenged ...
See moreThe rapid advancement of Generative Artificial Intelligence presents a paradigm shift for the architectural discipline by offering capabilities for form generation and visual exploration. However the integration of these technologies into the architectural workflow remains challenged by opacity and a lack of semantic understanding. This thesis investigates the role of AI not as an automation tool but as a co creative partner in the early stages of architectural ideation. Positioned at the intersection of Human Computer Interaction and Computational Design this research adopts a Research through Design methodology to systematically explore the potential of intelligent design tools. The study is grounded in the development of five distinct prototypes which evolve from two dimensional image generation to complex three dimensional form reconstruction. These systems serve as investigative probes to map the design space by testing specific capabilities such as latent space navigation and the translation of sketches into tectonic structures. The research synthesizes findings to distinguish between synthetic reasoning as the capacity for vast pattern recognition by the machine and human reasoning which encompasses tacit knowledge and contextual judgment. The thesis argues that successful integration requires a division of labor wherein the machine acts as an expander of the solution space while the human designer retains the critical role of goal formulation and semantic filtering. Ultimately this work contributes a theoretical and practical framework for architectural design augmented by machine intelligence. It articulates workflows that leverage computational strengths while mitigating limitations such as hallucination. By redefining the ontological status of AI from an autonomous generator to a collaborative agent this thesis offers a roadmap for a resilient and creatively enriched architectural practice.
See less
See moreThe rapid advancement of Generative Artificial Intelligence presents a paradigm shift for the architectural discipline by offering capabilities for form generation and visual exploration. However the integration of these technologies into the architectural workflow remains challenged by opacity and a lack of semantic understanding. This thesis investigates the role of AI not as an automation tool but as a co creative partner in the early stages of architectural ideation. Positioned at the intersection of Human Computer Interaction and Computational Design this research adopts a Research through Design methodology to systematically explore the potential of intelligent design tools. The study is grounded in the development of five distinct prototypes which evolve from two dimensional image generation to complex three dimensional form reconstruction. These systems serve as investigative probes to map the design space by testing specific capabilities such as latent space navigation and the translation of sketches into tectonic structures. The research synthesizes findings to distinguish between synthetic reasoning as the capacity for vast pattern recognition by the machine and human reasoning which encompasses tacit knowledge and contextual judgment. The thesis argues that successful integration requires a division of labor wherein the machine acts as an expander of the solution space while the human designer retains the critical role of goal formulation and semantic filtering. Ultimately this work contributes a theoretical and practical framework for architectural design augmented by machine intelligence. It articulates workflows that leverage computational strengths while mitigating limitations such as hallucination. By redefining the ontological status of AI from an autonomous generator to a collaborative agent this thesis offers a roadmap for a resilient and creatively enriched architectural practice.
See less
Date
2026Rights statement
The author retains copyright of this thesis. It may only be used for the purposes of research and study. It must not be used for any other purposes and may not be transmitted or shared with others without prior permission.Faculty/School
The University of Sydney School of Architecture, Design and PlanningAwarding institution
The University of SydneyShare