Söylem ve Mimari Ürün: Yapay Zekâ Destekli Karşılaştırmalı Bir Analiz Modeli

Yıl 2025, Cilt: 9 Sayı: 2, 223 - 231, 20.09.2025

Betül Uyan , Tayibe Seyman Güray

https://doi.org/10.54864/planarch.1680817

Öz

Mimari söylemin somut bir tasarım ürününe dönüşüm süreci, mimarlık kuramı ve tasarım
araştırmaları açısından her zaman ilgi çekici bir konudur. Tarihsel süreç boyunca toplumsal
olgular, mimari söylemi biçimlendirmiş; bu söylem de yapı ürünlerinin ortaya çıkışını
etkilemiştir. Ancak söylem ile tasarım ürünü arasındaki ilişki çoğu zaman belirsizdir. Genellikle
tasarımın erken aşamalarında geliştirilen söylem, projeye dair soyut değerleri barındırır ve bu
değerler metinsel ya da görsel ifadelerle temsil edilir. Tasarım bilişi alanındaki çalışmalar, bu
dönüşüm sürecini el-zihin etkileşimi üzerinden, sınırlı sayıda denemenin yapıldığı kapsamlı
süreçlerle incelemektedir. Bu çalışma, yapay zekâ destekli bir metinden-görüntüye modelin,
erken tasarım sürecine katkı sağlayıp sağlayamayacağı ve söylem-ürün ilişkisinin sorgulanabilir
doğasına yönelik bir değerlendirme yöntemi sunup sunamayacağı sorusuna odaklanmaktadır.
Çok sayıda yineleme yapabilme avantajıyla bu çalışma, söylem ve tasarım ürünü arasındaki
tutarlılığı test etmeyi amaçlamaktadır. Kuramsal çerçeve oluşturulduktan sonra, söylem ve
ürün ilişkileri bakımından farklı karakterde iki mimar ve yapıtı seçilmiştir: Rem Koolhaas’ın
MoMA Genişleme Yarışması için sunduğu öneri ve Brian Cantley’nin Syntaxonome adlı
çalışması. Yapay zekâ modeliyle elde edilen görseller, orijinal tasarımlarla karşılaştırılmış ve
söylem-ürün ilişkisi grafiksel ve sayısal oranlarla değerlendirilmiştir. Bu yöntem, mimarlık
kuramında nitel bir sorunsalın nicel olarak ele alınmasına yönelik yeni bir yaklaşım
sunmaktadır.

Anahtar Kelimeler

Mimari söylem , üretken YZ araçları , metinden görüntüye , insan-YZ işbirliği

Discourse and Architectural Artifact: An AI powered Comparative Analysis Model

Öz

The transformation of architectural discourse into tangible design artifacts remains a central
concern within architectural theory and design research. Historically, social phenomena have
continuously shaped architectural discourse, which in turn has informed the emergence of
built forms. However, the nature of the relationship between discourse and artifact remains
ambiguous. Typically developed during the preliminary design phase, discourse encapsulates
the intangible values of a project, often articulated through textual and visual media. Design
cognition research investigates this transformation, focusing on the interaction between hand
and mind via sketches, though such studies are limited by the number of iterations they can
analyze. This study explores whether artificial intelligence, specifically text-to-image
generation models, can offer a novel approach to evaluating the consistency between
architectural discourse and its resulting artifact. By enabling numerous iterations, the
proposed AI-supported method seeks to quantify a fundamentally qualitative research problem
in architectural theory. Two case studies were selected based on contrasting discursive and
formal characteristics: Rem Koolhaas’s Charette Submission for the Expansion of MoMA, and
Brian Cantley’s Syntaxonome. The discourse associated with each architect was input into a
text-to-image model to generate visual outputs. These were then compared with the original
artifacts to evaluate the alignment between discourse and design. Results were analyzed
through quantitative ratios and visual charts, offering a new perspective on the discourse-
artifact continuum and highlighting the potential of AI tools in early-stage architectural design.

Anahtar Kelimeler

Architectural discourse , generative AI , human-AI collaboration , text-to-image

Kaynakça

• Alberti, L. B. (1991). On the Art of Building in Ten Books (J. Rykwert, N. Leach, & R. Tavernor, Eds.). The MIT Press.
• Amer, F., Jung, Y., & Golparvar-Fard, M. (2023). Construction schedule augmentation with implicit dependency constraints and automated generation of lookahead plan revisions. Automation in Construction, 152, 104896–104896. https://doi.org/10.1016/j.autcon.2023.104896 2023.104896
• Baduge, S. K., Thilakarathna, S., Perera, J. S., Arashpour, M., Sharafi, P., Teodosio, B., … Mendis, P. (2022). Artificial intelligence and smart vision for building and construction 4.0: Machine and deep learning methods and applications. Automation in Construction, 141, https://doi.org/10.1016/j.autcon.2022.104440
• Cantley, B. (2013). Two Sides of the Page: The Antifact and the Artefact. Architectural Design, 83(5), 36–43. doi:10.1002/ad.1660 https://doi.org/10.1002/ad.1660
• Cantley, B. (2023). Speculative Coolness: Architecture, Media, the Real, and the Virtual. Taylor & Francis.
• Castro Pena, M. L., Carballal, A., Rodríguez-Fernández, N., Santos, I., & Romero, J. (2021). Artificial intelligence applied to conceptual design. A review of its use in architecture. Automation in Construction, 124, https://doi.org/10.1016/j.autcon.2021.103550
• Chaillou, S. (2022). Artificial Intelligence and Architecture: From Research to Practice. Birkhäuser.
• Chan, Y. H. E., & Spaeth, A. B. (2020). Architectural visualisation with conditional generative adversarial networks (cGAN). What machines read in architectural sketches. Ecaade 2020: Anthropologic-Architecture and Fabrication in the Cognitive Age, 299-308.
• Dorrian, M., Eeckhout, R., & Hendrickx, A. (Eds.). (2022). Drawing Architecture: Conversations on Contemporary Practice. Lund Humphries.
• Evans, R. (1997). Translations from Drawing to Building and Other Essays. The MIT Press.
• Goel, V. (1992). Ill-Structured Representations for Ill-Structured Problems. Presented at the Fourteenth Annual Conference of the Cognitive Science Society.
• Goldschmidt, G. (2014). Design Creativity. In Linkography. doi:10.7551/mitpress/9455.003.0008
• Goldschmidt, G. (2019). Disciplinary Knowledge and the Design Space (N. Börekçi, D. Koçyıldırım, F. Korkut, & D. Jones, Eds.). METU Department of Industrial Design.
• Gozalo-Brizuela, R., & Garrido-Merchan, E. C. (2023). ChatGPT is not all you need. A State of the Art Review of large Generative AI models. arXiv preprint arXiv:2301.04655. https://doi.org/10.48550/arXiv.2301.04655
• Karahan, H. G., Aktaş, B., & Bingöl, C. K. (2023, July). Use of language to generate architectural scenery with AI-powered tools. In International Conference on Computer-Aided Architectural Design Futures (pp. 83-96). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-37189-9_6
• Kim, H., Kang, H., Choi, H., Jung, D., & Hong, T. (2023). Human-building interaction for indoor environmental control: Evolution of technology and future prospects. Automation in Construction, 152, https://doi.org/10.1016/j.autcon.2023.104938
• Ko, J., Ennemoser, B., Yoo, W., Yan, W., & Clayton, M. J. (2023). Architectural spatial layout planning using artificial intelligence. Automation in Construction, 154, https://doi.org/10.1016/j.autcon.2023.105019
• Koolhaas, R. (1978a). Dali & Le Corbusier: the paranoid-critical method. Architectural Design, 48(2–3).
• Koolhaas, R. (1978b). Delirious New York. Oxford University Press.
• Leach, N. (2018). Design in THE Age of Artificial Intelligence. Landscape Architecture Frontiers, 6(2), 8–8. doi:10.15302/J-LAF-20180202
• Paananen, V., Oppenlaender, J., & Visuri, A. (2024). Using text-to-image generation for architectural design ideation. International Journal of Architectural Computing, 22(3), 458-474. https://doi.org/10.1177/14780771231222783
• Pallasmaa, J. (2012). The Eyes of the Skin: Architecture and the Senses. John Wiley & Sons Inc.
• Pérez Gómez, A., & Pelletier, L. (1997). Architectural representation and the perspective hinge.(No Title).
• Pérez-Gómez, A. (2016). Attunement: Architectural meaning after the crisis of modern science. MIT Press. https://doi.org/10.7551/mitpress/10703.001.0001
• Sawyer, R. K. (2022). The Dialogue of Creativity: Teaching the Creative Process by Animating Student Work as a Collaborating Creative Agent. Cognition and Instruction, 40(4), 459–487. https://doi.org/10.1080/07370008.2021.1958219
• Schumacher, P. (2011). The Autopoiesis of Architecture. John Wiley & Sons.
• Seyman-Guray, T. (2023). Investigating AI Applications in Construction Industry: A Systematic Review. Presented at the 6th International Conference of Contemporary Affairs in Architecture and Urbanism – Full paper proceedings of ICCAUA2023, 14-16 June 2023. https://doi.org/10.38027/iccaua2023en0153
• Sherkat, S., Garmaroodi, A. A., Wortmann, A., & Wortmann, T. (2023). Residential complex design as a Constraint Satisfaction Problem. Automation in Construction, 154, 104995–104995. https://doi.org/10.1016/j.autcon.2023.104995
• Somer, P. M. (2015). Metinden görsele mimaride ekfrasis. [Doktora Tezi, İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü].
• Spiller, N. (2018a). Magic Craftsmanship: A Glimpse into Bryan Cantley’s Thirdspace. Architectural Design, 88(2), 92–97. https://doi.org/10.1002/ad.2285
• Spiller, N. (2018b). That was Then, This is Now and Next. Architectural Design, 88(2), 6–15. doi:10.1002/ad.2273
• Thampanichwat, C., Wongvorachan, T., Sirisakdi, L., Chunhajinda, P., Bunyarittikit, S., & Wongmahasiri, R. (2025). Mindful Architecture from Text-to-Image AI Perspectives: A Case Study of DALL-E, Midjourney, and Stable Diffusion. Buildings, 15(6), 972. https://doi.org/10.3390/buildings15060972
• Tschumi, B. (1981). Manhattan Transcripts. MIT Press.
• URL-1. Dall-e, text to image model, retrieved from https://app.ai-pro.org/dream-photo?upid=4734i16mcLUBsNngBaTF (last access 21.04.2025)
• URL-2. Midjourney, generative artificial intelligence program and service, retrieved from https://www.midjourney.com/explore?tab=top (last access 21.04.2025)
• URL-3. Microsoft Bing, AI image generator, retrieved from https://www.bing.com/images/create?form=FLPGEN (last access 20.04.2025)
• URL-4. Adobe Firefly, generative AI powered features for creating image content, retrieved from https://www.adobe.com/learn/firefly/web/introduction-to-firefly (last access 21.04.2025)
• URL-5. Leonardo AI, AI image generator, retrieved from https://app.leonardo.ai/image-generation (last access 19.04.2025)
• URL-6 Stable Diffusion, text to image diffusion model, retrieved from https://stablediffusionweb.com/app/image-generator (last access 19.04.2025)
• URL-7 Openart, AI image generator, retrieved from https://openart.ai/create (last access 20.04.2025)
• URL-8 Architectural Design (AD), retrieved from https://archdesignjournal.com (last access 21.04.2025)
• Vesely, D. (1978). Surrealism and Architecture. Architectural Design, 48(2–3)