AI in Architecture: The Tools Changed. Talent Still Leads.
Part of our Architectural Intelligence Series
A computer can draw a perfect line. It cannot decide whether the line belongs there.
That distinction has shaped architecture through every major technological change, from hand drafting to AutoCAD, digital modeling, advanced visualization, and now artificial intelligence. Each tool has made the work faster, more precise, or easier to communicate. None has removed the need for an architect who can see the idea before the tool can express it.
As Meltzer/Mandl marks its 30th anniversary, the firm has lived through much of that transformation firsthand.
For co-founder Marvin Meltzer, architectural talent was once closely tied to the ability to draw. He came to the profession through painting and sketching, at a time when those skills were often treated as a basic qualification.
“If you went into architecture, it was because you could draw,” he says.
Marvin developed designs on tracing paper, working through one version after another. Rooms moved. Proportions changed. Each layer brought the project closer to the idea he had in mind.
“The more paper I used, the better I did.”
The paper was not simply a way to present the design. It was where Marvin tested it, challenged it, and found it.
When the Work Moved to the Computer
Computer drafting changed architectural practice quickly.
Tasks that once required hours of repetitive work could be completed in a fraction of the time. If ten apartments contained the same kitchen, the architect no longer had to draw that kitchen ten times. The team could draw it once, repeat it accurately, and revise each instance consistently.
Dimensions became more reliable. Changes became easier to incorporate. Renderings that once required hours of drawing and painting could be produced, adjusted, and shared digitally.
Marvin recognized the benefits, but the transition raised a concern.
“Where does the design thinking process enter?”
If drawing and thinking had once happened together on paper, architects now had to determine how the thinking would carry over to the screen.
For some, the design process moved directly into the computer. Marvin took a different route.
He never learned to use the computer as his primary design tool. He continued to sketch, review plans, identify problems, and direct revisions. His team translated those decisions into digital drawings, where the work could be measured, tested, and developed with greater precision.
His question was not, “How do I become the person operating the software?”
It was, “How do I use the computer to help me?”
That approach remains relevant. An architect does not need to personally operate every tool. The architect needs to understand the project, recognize what is not working, and communicate a clear direction to the people who can carry it forward.
The computer can make the change. Talent determines which change should be made.
Technology Gave Talent More Ways to Work
The rise of digital tools also changed the definition of who could become a strong architect.
When hand drawing was treated as essential, people with different strengths could be overlooked. An architect might understand planning, proportion, construction, circulation, or spatial experience without being a gifted illustrator.
Technology gave those abilities another way to reach the work.
Marvin points to his son and partner, Eli Meltzer, as an example. Eli did not need to approach architecture through painting or polished hand drawing to be comfortable and effective as an architect. Digital tools allowed a different set of strengths to come through.
That did not make talent less important. It made architectural talent broader and easier to express.
One architect may discover a design through tracing paper. Another may develop it through a digital model. A principal may establish the concept while a team uses software to test and refine it.
“If you are talented, you can do it no matter what,” Marvin says.
A talented architect learns how to work with the tools available, whether that means operating them directly or directing a team that does.
Technology does not create ability.
Efficiency Does Not Equal Quality
The greatest risk is not that technology will replace architects. It is that speed and polish will be mistaken for architectural quality.
“Efficiency does not give you quality,” Marvin says.
A computer can produce an accurate plan, but accuracy alone does not make the plan work. A rendering can look photographic, but visual realism does not resolve questions of proportion, context, circulation, or use. Software can generate multiple options, but it cannot determine which one is worth developing without informed direction.
“You still need to come up with ideas.”
That is the part technology cannot supply.
Good architecture depends on someone who can understand the problem, imagine a response, and recognize the difference between an efficient solution and a meaningful one. The tools can help test that response, correct it, repeat it, and communicate it. They can help talent move more quickly and work more precisely.
But the talent has to be there first.
Thirty Years of Better Tools
Over Meltzer/Mandl’s 30 years in practice, architectural tools have become faster, more accurate, and more capable.
The firm has adapted by using technology where it strengthens the work: to reduce repetition, improve precision, develop drawings, study alternatives, and communicate ideas. The tools continue to evolve, and AI will create new efficiencies and new ways to explore.
The principle remains the same.
Technology is most valuable in the hands of people who know what they are trying to achieve.
Or, as Marvin puts it, the work still has to be “in your head and heart” before you can deliver it.
The tools help talent reach the building.
They do not replace the talent that puts the idea there.
Comments