After Doctors, Mount Sinai Brings Ambient AI to Nurses

Nurses were spending more time behind a workstation than at the bedside. Mount Sinai Medical Center in Miami Beach is betting AI that listens can fix that.

September 09, 2026
After Doctors, Mount Sinai Brings Ambient AI to Nurses Health

Summary: Mount Sinai Medical Center in Miami Beach is rolling out Epic's Chart with Art ambient AI documentation tool to inpatient nurses, becoming among the first health systems in the nation to extend the technology beyond physicians, who saw a 30% reduction in charting time per patient encounter. Chief Nursing Officer Wendy Stewart said nurses had to learn to narrate assessments aloud since much nursing care happens without spoken conversation, starting with predictable head-to-toe assessments before expanding to pain reassessments and skin documentation. The tool transcribes in both English and Spanish, a meaningful capability given more than 30% of Mount Sinai's patients speak Spanish as their primary language, and every AI-generated note still requires nurse review before entering the medical record.

mountsinaimiamiambientainursingtechmiamibeachhealthcareaiepicsystems

Nurses were spending more time behind a workstation than at the bedside. Mount Sinai Medical Center in Miami Beach is betting AI that listens can fix that.

The health system is rolling out Epic's Chart with Art ambient documentation tool to inpatient nurses. It follows a successful pilot with physicians that cut charting time by 30% per patient encounter.

Why Nursing Required a Different Approach

Ambient AI works differently for nurses than for doctors. A physician visit captures a spoken conversation naturally. Much of nursing care doesn't involve talking out loud at all.

"We were finding that our nurses were spending a lot of time behind the workstation instead of really in front of their patients," said Wendy Stewart, Mount Sinai's chief nursing officer.

Nurses had to learn an entirely new habit: narrating their own work. They bring the tool into a patient's room, explain how it works, then verbalize findings while performing the assessment.

"That was probably one of the biggest learning curves for the staff," Stewart said.

Starting Small, With a Plan to Expand

Mount Sinai began with head-to-toe assessments specifically because they're relatively predictable. Pain reassessments and skin documentation are next.

Nurses helped design the workflow themselves, developing and practicing scripts before using the technology on real patients. Nursing leaders and educators trained in Epic's simulation environment first.

The pilot started with eight or nine day-shift nurses on a single medical-surgical unit. Plans call for expanding to roughly 20 nurses before rolling the tool out further.

A Language Capability That Matters Locally

The tool transcribes in both English and Spanish. That's a meaningful detail for Mount Sinai specifically, where more than 30% of patients speak Spanish as their primary language.

Because nurses narrate what they're checking out loud, patients hear more about their own exam than they typically would. Stewart said that's already prompting patients to ask follow-up questions about findings they wouldn't otherwise have known were being evaluated.

"This really has helped with the communication with nurses and their patients, for sure," Stewart said.

The Guardrail Built Into Every Note

Every AI-generated note still requires a nurse's review and correction before it becomes part of the official medical record.

"People are much more willing to embrace technology when they view it as a value add, as opposed to a real change to their workflow," said Dr. Alon Weizer, Mount Sinai's chief medical officer.

What Comes Next for Miami's Nursing Staff

Night-shift nurses are next in line, giving Mount Sinai a chance to test the tool in a different workflow entirely. Intensive care and labor and delivery units could eventually follow.

Stewart's advice to other health systems considering the same move is simple. Involve nursing staff in building the workflow before the technology arrives, not after the decisions are already made.

Bascom Palmer Spinout Heru Cuts Eye Pretesting to 90s