UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

AI-Driven Pedagogies in Architecture: A framework for early-stage design education

Ponzio, Angelica; Fatah gen. Schieck, Ava; (2025) AI-Driven Pedagogies in Architecture: A framework for early-stage design education. In: Proceedings of theSIGraDi 2025: META RESPONSIVE APPROACHES. (pp. pp. 161-172). Sociedad Iberoamericana de Gráfica Digital (SIGraDi)

[thumbnail of ID_259_Final 23Sept25.pdf] Text
ID_259_Final 23Sept25.pdf - Published Version
Access restricted to UCL open access staff until 18 February 2026.

Download (1MB)

Abstract

This exploratory study investigates the role of 2D generative Artificial Intelligence (GenAI) applied during early design-stages in architectural education settings. Grounded in Donald Schön’s (1983) reflective practice and adopting a DesignBased Research (DBR) methodology, it presents frameworks where AI functions as a conversational design partner. Three exploratory phases are presented: (1) a theoretical stage—aimed at building familiarity with tools and refining processes; (2) a test stage— a workshop testing the pipeline; and (3) a design studio, employing a structured AI matrix, to explore the concept of parameter-based prompt blending across multimodal platforms. Initial findings indicate that GenAl enhances conceptual stages when anchored in clear design intentions. Moreover, the recent emergence of multi-model AI ecosystems enabled cyclical and layered workflows through Human-AI dialogues producing outputs that transcend generic architectural representation. Furthermore, these systems also foster new forms of pedagogical dialogue between educators and students within the academic environment.

Type: Proceedings paper
Title: AI-Driven Pedagogies in Architecture: A framework for early-stage design education
Event: SIGraDi 2025: META RESPONSIVE APPROACHES
Location: Buenos Aires, Argentina
Dates: 19 Nov 2025 - 21 Nov 2025
ISBN-13: 978-9915-9635-3-2
Publisher version: https://sigradi.org/sigradi2025/
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: Artificial Intelligence, Multimodal Generative Models, Architectural Design, Early-Stage Ideation, Parameter Prompt Blending
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment > The Bartlett School of Architecture
URI: https://discovery.ucl.ac.uk/id/eprint/10218910
Downloads since deposit
1Download
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item