AI in architecture: translating disruptive ideas into something buildable
Francisco Piqueras
Talking today about AI in architecture It's no longer science fiction… although at IA Manufacturing we've had that "AI" in our name from the very beginning. It wasn't planned—or so we say—but it's curious how those same initials that have always been with us have become key in the present of architecture.
Artificial intelligence began by appearing tentatively in talks, design tools, competitions… and Without us realizing it, it became part of the studio's daily routine. Not with grandiose promises, but with concrete solutions that help to better plan projects.
Between fascination and professional judgment
Whenever a disruptive technology emerges, it's normal for doubts to arise. In our study, the AI in architecture It aroused both enthusiasm and caution. To what extent can a machine-generated proposal be trusted? How does an architect's intuition fit into a data-driven process?
Questions like these They are not a barrier, but part of the dialogue we have opened. in the team, where both innovation and the judgment that experience provides are valued.
Not everything that glitters is an algorithm
Over time, we learned to look beyond the glitz of the new. Yes, there are tools that are astonishing in their ability to generate complex ideas in seconds. But we also understood that, Without a critical eye, those results may remain superficial..
That's why, at IA Manufacturing, we're interested in how artificial intelligence can complement —not replace— the creative process, helping to detect patterns, anticipate conflicts, or open paths that the architect then explores with his own logic and sensitivity.
Models as a physical anchor for using AI in architecture
At IA Manufacturing we experience it every day: artificial intelligence applied to architecture It allows you to explore hundreds of variations of the same design in a matter of minutes.. But that digital freedom needs to be anchored in something tangible, something that can be seen, touched, and shared.
That's where the models They remain fundamental. They are the connection point between the proposal generated by artificial intelligence and the real spatial perception of the design team or the end client.
Seeing is understanding… and verifying that it holds up
In the digital environment everything is possible: impossible structures, floating shapes, angles that defy gravity. But outside of the rendering, the world has rules.
Models help to test those ideas that work perfectly in 3D, but need adjustment in reality. They allow us to detect errors, validate proportions, and understand if what looks logical on screen can actually be built.. They are that necessary bridge between digital creativity and physical feasibility.
Lessons learned from real-world projects using AI in architecture
There are projects in which AI has ceased to be a one-off tool and has become a permanent ally. We've seen it in commissions where the parametric design It generated hundreds of structural variants in a matter of minutes, or in sun exposure studies that cross-referenced data on orientation, climate, and urban environment.
In those cases, artificial intelligence did not replace human work, but rather enhanced it: It allowed for more informed decisions to be made from the beginning.
The model as final validation
In all these processes, the physical model continues to play a key role. No matter how many simulations are run or how accurate the data is: When the project is built at scale, new questions arise.. Is the flow clear? Does the volume breathe well within its context? Does the structure make sense beyond the algorithm?
At IA Manufacturing, whenever we use AI in a project, there's a physical model that helps us confirm—or adjust—what seemed resolved digitally. These models don't necessarily have to be aesthetically pleasing; sometimes they're parts of a structure, and other times they're surfaces and finishes that... We need to experiment before they are brought into reality.
Open questions and a shared future
The AI in architecture It's still taking its first steps, but Its impact is already being felt.. How far will it go? Who will make the final decisions in a project: the algorithm, the architect, or a collaboration between both?
At IA Manufacturing we believe that The important thing is not to find definitive answers, but to be willing to continue asking ourselves questions.. Artificial intelligence poses ethical, technical, and creative challenges, and that's where studies that know how to adapt—without losing their criteria—make the difference.
A new role for the model maker
In this changing context, the role of model maker It also evolves. It is no longer just about building scale models, but about translating complex digital ideas into understandable objectsverifiable and useful. The model maker becomes a bridge between worlds: between the virtual and the real, between the projected and the possible. And that, in times of artificial intelligence, is more necessary than ever.
What if we let AI think for us?
Every advance in AI applied to architecture brings with it an opportunity… and also a responsibility. Because yes, algorithms can suggest efficient solutions, optimize resources, and even generate forms we wouldn't have imagined. But we also run the risk of delegating too much.. What happens if we let the machine decide for us? Where does that leave human judgment, experience, and intuition?
At IA Manufacturing we believe that Artificial intelligence should be a tool, not a superior voice.. We're more interested in what it allows than what it imposes. That's why we keep asking questions, designing models, and closely observing the process. Because thinking is still what we do. And that—for now—no one else can do for us.
Did you enjoy this article?
Want to see other similar posts?
