This thesis investigates the role of generative artificial intelligence in architectural design and asks what becomes of the designer's agency as probabilistic systems are brought into the design process. It addresses three difficulties limiting their adoption: the loss of control associated with opaque, black-box models; the gap between the raster imagery these tools generate and the vector precision of architectural documentation; and the absence of a structured way to evaluate a tool set that changes rapidly. Against these, it categorizes generative tools by function, tests workflows that balance creative exploration with geometric control, and charts a route for AI-assisted design. The approach is Research through Design, combining literature review, technical analysis, experimental testing, and design practice. Generative tools were first sorted by modality and interface, then assessed across the conceptual, schematic, and realization phases. Comparative testing measured image-generation and image-editing models against four criteria: usability, prompt adherence, consistency, and versatility. These findings were consolidated into an evaluation framework and examined through a case study of an unusual typology, a contemporary Islamic center. The study finds generative AI most effective as an augmentative partner rather than an autonomous designer. Current models excel at concept exploration, visualization, and rapid iteration, yet remain unreliable for orthographic documentation and precise technical work. The most productive workflows pair human judgment with progressively constrained generation, turning stochastic output into a controlled instrument. On authorship the thesis reaches no final verdict, showing only that it is held or lost according to how the engagement is structured. Its enduring contribution is methodological: a transferable framework for bringing generative AI into practice on terms the architect sets, with control and intent kept in view.
Questa tesi indaga il ruolo dell'intelligenza artificiale generativa nella progettazione architettonica e si chiede cosa accada all'agency del progettista quando sistemi probabilistici entrano nel processo. Affronta tre difficoltà che ne limitano l'adozione: la perdita di controllo legata a modelli opachi a scatola chiusa; il divario tra le immagini raster prodotte da questi strumenti e la precisione vettoriale della documentazione architettonica; e l'assenza di un metodo strutturato per valutare un insieme di strumenti in rapido mutamento. A fronte di ciò, classifica gli strumenti generativi per funzione, sperimenta flussi di lavoro che bilanciano esplorazione creativa e controllo geometrico e traccia un percorso per il progetto assistito dall'IA. L'approccio è il Research through Design e combina revisione della letteratura, analisi tecnica, sperimentazione e pratica progettuale. Gli strumenti generativi sono stati prima ordinati per modalità e interfaccia, poi valutati nelle fasi concettuale, schematica e di realizzazione. Test comparativi hanno valutato modelli di generazione e di editing rispetto a quattro criteri: usabilità, aderenza al prompt, coerenza e versatilità. I risultati sono confluiti in un quadro di valutazione, esaminato in un caso studio di una tipologia inconsueta, un centro islamico contemporaneo. Lo studio rileva che l'IA generativa è più efficace come partner aumentativo che come progettista autonomo. I modelli attuali eccellono nell'esplorazione concettuale, nella visualizzazione e nell'iterazione rapida, ma restano inaffidabili per la documentazione ortografica e il lavoro tecnico di precisione. I flussi di lavoro più produttivi uniscono il giudizio umano a una generazione progressivamente vincolata, trasformando l'output stocastico in uno strumento controllato. Sull'autorialità la tesi non giunge a un verdetto definitivo, mostrando solo che è mantenuta o perduta secondo come l'interazione è strutturata. Il contributo duraturo è metodologico: un quadro trasferibile per introdurre l'IA generativa nella pratica alle condizioni stabilite dall'architetto, mantenendo in vista controllo e intento.
Latent architecture: a journey through the control of generative AI's hallucinations
Arafa, Ahmed Tareef Arafa Mohamed
2025/2026
Abstract
This thesis investigates the role of generative artificial intelligence in architectural design and asks what becomes of the designer's agency as probabilistic systems are brought into the design process. It addresses three difficulties limiting their adoption: the loss of control associated with opaque, black-box models; the gap between the raster imagery these tools generate and the vector precision of architectural documentation; and the absence of a structured way to evaluate a tool set that changes rapidly. Against these, it categorizes generative tools by function, tests workflows that balance creative exploration with geometric control, and charts a route for AI-assisted design. The approach is Research through Design, combining literature review, technical analysis, experimental testing, and design practice. Generative tools were first sorted by modality and interface, then assessed across the conceptual, schematic, and realization phases. Comparative testing measured image-generation and image-editing models against four criteria: usability, prompt adherence, consistency, and versatility. These findings were consolidated into an evaluation framework and examined through a case study of an unusual typology, a contemporary Islamic center. The study finds generative AI most effective as an augmentative partner rather than an autonomous designer. Current models excel at concept exploration, visualization, and rapid iteration, yet remain unreliable for orthographic documentation and precise technical work. The most productive workflows pair human judgment with progressively constrained generation, turning stochastic output into a controlled instrument. On authorship the thesis reaches no final verdict, showing only that it is held or lost according to how the engagement is structured. Its enduring contribution is methodological: a transferable framework for bringing generative AI into practice on terms the architect sets, with control and intent kept in view.| File | Dimensione | Formato | |
|---|---|---|---|
|
2026_07_Arafa.pdf
accessibile in internet per tutti
Descrizione: Thesis Text
Dimensione
69.39 MB
Formato
Adobe PDF
|
69.39 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/260483