By Charles Michelson, AIA, ACHA, LEED AP
In four decades of designing healthcare environments, I have watched care itself migrate steadily out of the hospital. Procedures that once demanded an overnight stay now happen in ambulatory surgery centers, imaging suites, and medical office buildings closer to where people live. Artificial intelligence arrives just as this shift accelerates, and it may prove most consequential not in the acute hospital, but in the distributed, high-volume world of outpatient care, because it changes not only how we draw these buildings but how they perform.
Today, AI is already reshaping the earliest and most difficult phase of our work: programming and planning. Generative tools can model thousands of layout options in moments, testing exam-room modules, staff travel distances, and clinic workflows against real scheduling and utilization data that once took months to analyze. Rather than replacing the architect’s judgment, these systems widen the field of possibilities we evaluate, helping us right-size a clinic to actual demand rather than to rules of thumb. Patient-flow simulation lets us stress-test an ambulatory surgery center or a busy imaging suite before a single wall is framed, exposing throughput bottlenecks while they remain inexpensive to fix. On the construction side, AI-driven estimating and progress tracking are compressing the schedules that make or break outpatient projects competing on speed to market.
The near future is more ambitious. Digital twins, living models fed by a facility’s own data, will let us observe a clinic network in operation and continuously refine it, closing the loop between design intent and operational reality across a whole hub-and-spoke system. Exam and procedure rooms are becoming responsive, adjusting lighting, acoustics, and setup to the visit at hand, while telehealth is folded into the physical plant rather than bolted on. Designing for this means planning the invisible as carefully as the visible, weaving sensors, networks, and computational infrastructure into flexible, convertible rooms that can change function as service lines evolve.
What excites me most is AI’s potential to return time to caregivers. In outpatient settings, where a provider may see dozens of patients a day, intelligent handling of documentation, translation, and routine monitoring frees clinicians for the human work only they can do, and our buildings must support that restored attention rather than compete with it. The best care environments have always been quietly therapeutic; AI gives us new instruments to make them responsive as well.
But intelligence is not a substitute for wisdom. These tools inherit the assumptions in their data, and a flawed dataset can encode a flawed workflow into a building meant to serve a community for decades. Privacy, equity of access, and clinical safety cannot be afterthoughts, and no algorithm should sign a drawing. Architects, engineers, and clinicians must remain accountable for what the models propose.
I remain optimistic. The measure of any tool is whether it helps us serve patients, and the people who care for them, more humanely. Used with discipline and humility, AI can help us design outpatient environments that are not only more efficient, but more accessible, adaptable, and kind. That has always been the goal; we simply have better means to reach it.
Charles Michelson, President, Saltz Michelson Architects, can be reached at (954) 266-2700 or cmichelson@saltzmichelson.com or visit www.saltzmichelson.com.
