Home Proceedings Program News Contact
PDF download
Cite article
Share options
Informations, rights and permissions
Issue image
Vol 4, 2025
Pages: 1345 - 1355
Research paper
Architecture Editor: Vuk Milošević
See full issue
Metrics and citations
Abstract views: 374
PDF Downloads: 43
Google scholar: See link
Article content
  1. Abstract
  2. Disclaimer
Published: 11.09.2025. Research paper Architecture Editor: Vuk Milošević

AI SUPPORTED PYTHON PROGRAMMING IN ARCHITECTURAL FORM-FINDING PROCESSES

By
Mirjana Devetaković Orcid logo ,
Mirjana Devetaković
Contact Mirjana Devetaković

Faculty of Architecture, University of Belgrade , Belgrade , Serbia

Jelena Ivanović Orcid logo ,
Jelena Ivanović

Faculty of Architecture, University of Belgrade , Belgrade , Serbia

Djordje Djordjević Orcid logo ,
Djordje Djordjević

Faculty of Architecture, University of Belgrade , Belgrade , Serbia

Djordje Mitrović Orcid logo
Djordje Mitrović

Faculty of Architecture, University of Belgrade , Belgrade , Serbia

Abstract

In recent years, artificial intelligence (AI) has become widely accessible, with powerful AI tools integrated into common ICT services and new AI-based applications emerging almost daily, often freely available. The Architecture, Engineering, Construction, and Operation (AECO) sector is no exception, as numerous studies highlight AI-enhanced applications supporting various aspects of the industry. While many focus on visualization, AI is also increasingly contributing to design and construction processes.

This paper explores the process of architectural form generation using Rhino 3D, leveraging Python scripting for parametric geometric modeling. ChatGPT-3 was utilized to generate Python scripts compatible with the Rhino 3D Python Editor, benefiting from its ability to integrate mathematical concepts with programming logic and respond to textual programming prompts. Through this approach, a series of complex geometric forms was modeled, visually examined, and selectively 3D printed.

Following an introduction to the relevant mathematical concepts and digital tools (ChatGPT, Python, and Rhino 3D), the paper details the methodology for guiding AI through parametric modeling. Since achieving the desired results is not always straightforward, specific strategies for controlled, purpose-driven interaction with ChatGPT are outlined. The study is illustrated with characteristic examples, and the generated architectural forms are analyzed, discussing the advantages and limitations of the approach.

The paper concludes by examining the evolving technological skillset required for contemporary architectural design in the AI era.

Funding Statement

This research has been funded by the Serbian Ministry of Science, Technological Development and Innovation, within the Project No. 200090 (University of Belgrade, Faculty of Architecture).

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.