CLASSIFICATION OF ARTIFICIAL INTELLIGENCE TOOLS IN ARCHITECTURE

Abstract

Purpose. The article seeks to address inconsistent AI terminology in architecture by proposing a concise multi axis classification aligned with the title and the stated problem. Four axes are defined and applied to tools across design, construction, and operations: functional and teleological purpose, information and algorithmic technique, process and lifecycle stage, and regulatory, ethical, resource and ecological factors. Methodology. A range of scientific methods was used to achieve the aim. A systematic literature review identified primary studies and practical cases under explicit inclusion criteria. Thematic coding organised the collected materials into clear themes and recurrent attributes, enabling analysis and description of AI tools in architecture by purpose, data and model type, lifecycle stage, and safety and ethics requirements, followed by integration into a unified matrix. Comparative analysis of practical examples, aligned with the four classification axes, revealed typical profiles for each tool and identified their strengths and weaknesses. Results. Four key axes of AI in architecture are identified: system purpose and functionality (e.g., generative design, optimization, analysis, automation); data and algorithm types (rule-based or knowledge-driven vs. machine learning, including supervised, unsupervised, reinforcement, and generative paradigms); integration stage in the AEC lifecycle (from conceptual design and planning through construction, operation, and decommissioning); compliance with safety, ethical, regulatory, and sustainability requirements (addressing bias mitigation, explainability, standards, energy efficiency). Identified data and illustrative tools (generative design software, construction robots, building monitoring systems) are mapped to the taxonomy. Scientific novelty. The study introduces an integrative classification approach that simultaneously accounts for the technical and teleological taxonomy of AI tools, their lifecycle deployment stage, and conformance to ethical, regulatory, and ecological criteria. Practical relevance. The results can help architects and AI developers to structure knowledge of existing solutions, implement principles of sustainable AI use in architectural practice.