The Designer in the Loop: Navigating Informed Creativity through Human-AI Collaboration
Year: 2026
Type: conference-paper
Abstract: While Large Language Models (LLMs) are reshaping computational design, their role in architectural education remains underexplored. This paper presents a study investigating how LLMs support "informed creativity" by analyzing participant outputs across four computational domains: 3D scene generation, interactive mini-games, 2D animations, and data visualization. By positioning the designer as an active cognitive collaborator rather than a passive user, the study examines how language-based steering influences design agency and learning curves. Results indicate that LLM-assisted workflows lower technical entry barriers and expand the speed and breadth of design iteration. However, technical limitations, such as hallucinated code and a lack of contextual awareness, underscore the strict necessity of informed human oversight (Bernstein, 2018). Ultimately, the paper argues for new architectural pedagogical frameworks that integrate LLMs as catalysts for exploration, effectively bridging the gap between conceptual intent and computational materialization.
Source: eCAADe proceedings
Author: Ozan Yetkin
Language: English
Subfield: Architecture
Field: Engineering
Domain: Physical Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7213894048
