By Regina Gonthier, UIA President
In today’s global and digital world, AI is being integrated into design and construction processes, influencing everything from concept and spatial analysis to documentation and performance. Artificial Intelligence is already shaping architectural practice, and, as in many fields, its development in architecture and urban planning is accelerating. It is essential to be aware of both its potential and its limitations.
At the 2024 Annual Meeting of the World Economic Forum, a key message was shared: “As AI innovation advances rapidly, we see its influence across the built environment in expected and unexpected ways, nowhere more so than in design”. The potential to change how buildings are used and operated, and how they connect with people and wider urban systems, was underlined. The possibilities seem endless.
In architecture, AI is being used to generate early concepts, optimise layouts, simulate environmental performance, and streamline repetitive tasks. In construction, it supports scheduling, material tracking, and maintenance. AI can increase efficiency, reduce costs, and improve precision across all design phases. It can assist with data analysis for climate-responsive solutions and help architects navigate increasingly complex regulatory and technical environments. Its ability to process vast datasets in real time opens new opportunities.
However, serious challenges and risks must also be addressed.
One major risk involves authorship and intellectual property. Architectural services are intellectual services, but AI-generated content raises unresolved questions about who holds the rights to the output—especially when tools rely on datasets that may include copyright-protected material.
Another risk involves transparency and accountability. Algorithms are not neutral. Decisions made by AI systems are shaped by the data they are trained on, which may contain biases—intentionally or not. If architects rely on AI without understanding its logic or assumptions, they risk perpetuating errors or poor design choices without being aware of it.
There is also the risk that architecture will become anonymised, ceasing to express cultural diversity. Many AI systems are programmed using data produced in specific cultural contexts. Applying them indiscriminately around the world risks erasing cultural, climatic, and social specificities.
Ethical standards must be central to this discussion. The professional judgment of architects must remain paramount in all processes involving design decisions. AI should not be approached as a neutral assistant. It is a system designed by others, trained on specific datasets, and programmed with certain priorities. These factors shape its results. It is therefore crucial to understand how these systems work before integrating them into workflows.
In 2023, The Guardian posed a fundamental question: “Will AI wipe out architects?”. Two years later, despite growing interest, adoption remains limited in practice.
The American Institute of Architects recently released a study showing that while 84% of architects in the United States are optimistic about AI’s potential to address complex challenges, only 6% reported using AI tools regularly in their work. At the same time, 90% expressed concerns about its implications.
This reflects a common situation in many parts of the world: awareness and interest are growing faster than actual use. Factors such as cost, access to training, and uncertainty about legal responsibilities slow down implementation.
Nonetheless, the impact on professional practice will be significant. Some human-made tasks will be performed by AI. At the same time, new roles and skills will emerge. The ability to understand, audit, and refine AI tools will become increasingly relevant. AI will inevitably cause deep social changes. Not all regions, offices, or professionals have equal access to technology or infrastructure. There is a serious risk that AI may deepen existing inequalities—between countries, between large firms and small practices, between those with technical support and those without, between those who can afford training and those who cannot.
Access to training must be ensured, enabling architects not only to use AI tools, but to evaluate their reliability, trace their outputs, and critically assess their relevance. Educational institutions must adapt curricula accordingly, and professional development must support mid- and late-career architects as well.
Beyond technical and legal questions, it is essential to maintain a clear sense of purpose. The use of AI must align with broader societal goals. Reducing emissions, improving resilience, strengthening social cohesion and cultural identity—these remain architectural priorities. Tools must serve these ends, not distract from them.
Architecture plays a critical role in addressing informal settlements, disaster risk, and social exclusion. In these contexts, the use of AI must be especially cautious. It must not displace local practices or overlook the lived experience of communities. Instead, it should be used—if and when appropriate—to support more accurate planning, improve resilience, and strengthen access to essential infrastructure and services, always under the guidance of human professionals who understand the context.
Architecture is not only about solving problems efficiently. It is about making choices about how we want to live together, how we relate to the natural environment, and how we build a future that is equitable and resilient.
AI can assist in this process, but it should not define it. Architects must maintain their specific role: understanding human needs, interpreting cultural contexts, and shaping spaces that respond to both.
In this context, artificial intelligence must remain a means, not an end.
Technology should serve humanity, not command it.
Architects must retain control and responsibility for their work. Local knowledge, traditional techniques, and community input must continue to influence design. AI is not just a tool—it is being developed to become an assistant, and perhaps even more than that: a partner that may hold its opinions. The key questions are: who decides, and who programmes? Can those responsible for programming be trusted? Can the information that nourishes these algorithms be verified? Algorithms must be moralised, but in an age where humanity is losing its fundamental values, the question arises: which ethics should guide them? Elaborating relevant legislation with appropriate restrictions will certainly help—but will it be respected? There is a real danger of misuse, potentially leading to totalitarianism.
The UIA has declared AI a priority topic for the 2023–2026 term. The Architectural Education Commission has been asked to include recommendations in the Education Charter. The Professional Practice Commission is working on recommendations for AI to be included in the International Standards of Professionalism in Architectural Practice. Ethical transparency and responsible use are fundamental requirements, alongside the protection of authors’ rights, maintaining the architect’s decision-making authority, and regulatory adaptation. This document is expected to be presented at the 2026 Congress and General Assembly in Barcelona.