Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
Design automation has consistently been a principal focus of computational design since its advent, particularly within the architectural field. Consequently, significant attention has been directed toward automation and optimization using numerically quantifiable design elements. The advent of AI in computational design, especially during the Generative Adversarial Network (GAN) era, extends beyond traditional methodologies, offering potential for a parallel design Intelligence that never existed before. Given that each existing, imaginary, and future iteration contributes to a potential seed of AI dataset, it is conceivable that every aspect of the design process can be automated, accelerated, and made optimization friendly. However, the potential of design-enabled synthetic intelligence raises a critical question: can AI-driven design achieve complete contextual coherence akin to the cases when a human designer is at the job, with comprehensive awareness of object-oriented, environmental, and scalability factors, inclusive of both human and non-human contextual elements?This research investigates the feasibility and challenges of developing a computational design system powered by synthetic intelligence using tools that are familiar in the Architecture, Engineering & Construction (AEC) industry. It examines the hurdles in creating an autonomous design co-pilot and interconnects the advancements in contemporary computational design tools with the mainstream AI development. Furthermore, this study projects how a new generation of hybrid systems could be structured, embodying a design-enabled AI-powered co-pilot.