From Clay to Code: Typological and Material Reasoning in AI Interpretations of Iranian Pigeon Towers

2025-12-29

arXiv.org

AI Summary

This study investigates the capacity and cognitive limitations of generative AI in understanding the typology and material logic of Iranian pigeon towers, a vernacular architectural form. Employing a three-stage prompting strategy—reference, adaptation, and speculation—across Midjourney v6, DALL-E 3, and Stable Diffusion XL, the authors develop a multidimensional evaluation framework assessing type, materiality, environmental response, realism, and cultural specificity. They propose a “computational vernacular reasoning” framework to distinguish between visual similarity and genuine architectural logic comprehension. Findings indicate that while AI accurately reproduces geometric forms, it frequently misinterprets material properties and climate-responsive design principles. Reference images enhance realism but constrain creative variation, whereas reference-free generation fosters formal exploration at the cost of cultural specificity, often resulting in ambiguous or generic outcomes.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

This study investigates how generative AI systems interpret the architectural intelligence embedded in vernacular form. Using the Iranian pigeon tower as a case study, the research tests three diffusion models, Midjourney v6, DALL-E 3, and DreamStudio based on Stable Diffusion XL (SDXL), across three prompt stages: referential, adaptive, and speculative. A five-criteria evaluation framework assesses how each system reconstructs typology, materiality, environment, realism, and cultural specificity. Results show that AI reliably reproduces geometric patterns but misreads material and climatic reasoning. Reference imagery improves realism yet limits creativity, while freedom from reference generates inventive but culturally ambiguous outcomes. The findings define a boundary between visual resemblance and architectural reasoning, positioning computational vernacular reasoning as a framework for analyzing how AI perceives, distorts, and reimagines traditional design intelligence.

Problem

• architectural intelligence
• vernacular architecture
• generative AI
• material reasoning
• cultural specificity

Innovation

• computational vernacular reasoning
• generative AI
• architectural intelligence
• material reasoning
• typological interpretation