GENERATIVE ARTIFICIAL INTELLIGENCE IN THE CONCEPTUAL MODELLING OF ARCHITECTURAL OBJECTS CAPABILITIES AND VERIFICATION CRITERIA

Generative artificial intelligence expands the tools available for conceptual modelling by producing textual descriptions, visual alternatives, layout schemes, program code, and geometric representations. The article aims to establish the conditions for using these outputs and to systematise criteria for checking them before further design development. The study employs a targeted analysis of scientific publications, comparison of official documentation and developer materials, comparison of output types, and methodological synthesis. Architectural imagery, spatial layouts, executable code, and models created by that code are distinguished. Two modes are identified: exploration that permits geometric changes and image processing that preserves specified spatial parameters. Eight criteria are assigned to the verification of mandatory requirements, validation of fitness for purpose, and comparison of admissible alternatives; evidence and an outcome status are specified for each check. The roles of ChatGPT in prompt preparation, Midjourney in visual exploration, Magnific in subsequent image processing, and Omegarender’s AlphaRender and VizMaker, described in VizAcademy materials, in specialised visualisation are explained. An AI-agent architecture with controlled access to 3ds Max and Archicad through application programming interfaces and adapters is proposed. The workflow is specified for 3ds Max with Corona Renderer: baseline rendering, generative processing, verification of changes, and reconstruction of the accepted design decision in the scene. The contribution is the alignment of generative output type, available evidence, and the decision on further use. Its practical significance lies in connecting generative exploration with professional modelling and architectural visualisation education.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24322925
Primary Topic
Architecture and Computational Design
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

GENERATIVE ARTIFICIAL INTELLIGENCE IN THE CONCEPTUAL MODELLING OF ARCHITECTURAL OBJECTS CAPABILITIES AND VERIFICATION CRITERIA

Роман Петрович Гончарик
3 links
Zenodo (CERN European Organization for Nuclear Research)
Architecture and Computational Design
article

GENERATIVE ARTIFICIAL INTELLIGENCE IN THE CONCEPTUAL MODELLING OF ARCHITECTURAL OBJECTS CAPABILITIES AND VERIFICATION CRITERIA

Роман Петрович Гончарик
article en

Abstract

Generative artificial intelligence expands the tools available for conceptual modelling by producing textual descriptions, visual alternatives, layout schemes, program code, and geometric representations. The article aims to establish the conditions for using these outputs and to systematise criteria for checking them before further design development. The study employs a targeted analysis of scientific publications, comparison of official documentation and developer materials, comparison of output types, and methodological synthesis. Architectural imagery, spatial layouts, executable code, and models created by that code are distinguished. Two modes are identified: exploration that permits geometric changes and image processing that preserves specified spatial parameters. Eight criteria are assigned to the verification of mandatory requirements, validation of fitness for purpose, and comparison of admissible alternatives; evidence and an outcome status are specified for each check. The roles of ChatGPT in prompt preparation, Midjourney in visual exploration, Magnific in subsequent image processing, and Omegarender’s AlphaRender and VizMaker, described in VizAcademy materials, in specialised visualisation are explained. An AI-agent architecture with controlled access to 3ds Max and Archicad through application programming interfaces and adapters is proposed. The workflow is specified for 3ds Max with Corona Renderer: baseline rendering, generative processing, verification of changes, and reconstruction of the accepted design decision in the scene. The contribution is the alignment of generative output type, available evidence, and the decision on further use. Its practical significance lies in connecting generative exploration with professional modelling and architectural visualisation education.

Zenodo (CERN European Organization for Nuclear Research)
Architecture and Computational Design
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon