Moxi at 25 US Hospitals: Why Scale Matters, and How to Check It

What does a 25 facility count actually measure?

The figure most often repeated for Moxi, the hospital delivery robot made by Diligent Robotics, is more than 25 US hospital facilities. CBS News reported that Diligent told it Moxi robots were at 25 hospitals nationwide. Nurse.org, citing company statistics, reports nearly 100 robots across more than 25 facilities and more than 1.25 million tasks completed to date.

Both figures come from the vendor side. We could not locate the July 2026 S&P Global article often cited for the count, so we do not attribute the number to that outlet. Treat it as a starting point for due diligence, not as proof of performance on any particular floor.

A facility count is not a robot count, and neither equals a unit that runs well every shift. On the reported numbers, a typical site averages roughly four robots. That is a thin bench for a site that can lose a unit to a charging fault, a sensor cleaning or a repair visit.

Counts also shift between reports. A later CBS piece describes roughly two dozen other hospitals and names Cedars-Sinai, so anyone quoting a figure internally should pin down the date and the definition behind it.

Why does fleet size affect support and spare units?

Support depth is where scale shows up first. A vendor running many sites has more engineers in more regions, more repair cases to learn from and more reason to keep spare modules on hand. A hospital with a single unit has a different problem. When that robot is down, the corridor runs on staff again.

Spare unit coverage belongs on the first call. Ask whether the vendor can place a loaner or backup robot during repairs, how fast an on site visit can happen, and whether remote triage can clear a fault before anyone walks to the supply closet. A backup robot program matters more in a hospital than in many warehouses, because a stalled medication transport run has a clinical face.

How does scale shape staff training?

Training scales with the number of people who need to be fluent. A pilot on one wing needs a small group of trained super users. A system adding robots across several campuses needs a repeatable curriculum for environmental services, nursing and transport staff, plus a way to train night shifts without relying on the day team.

A vendor with this reported reach has had to build onboarding that can repeat from site to site, at least in principle. The test is local, though. Ask who trains the trainers, how long a new unit takes to reach steady use, and what happens when staff turnover thins out the people who know the robot's habits.

Why check a count through named reference sites?

A vendor count tells you what the vendor chooses to report. Named reference sites let you check whether units still run a year after go live. The counterexample matters here. Nurse.org reports that MultiCare, a Washington hospital system, discontinued a program of 14 Moxi robots after it did not deliver the expected results. One report is not a verdict on the product, but it shows why a reference call should cover failures as well as wins.

Children's Hospital Los Angeles has been described as a flagship site, and CBS coverage names Cedars-Sinai among the hospitals. Use these as leads to call, not as endorsements of performance. Ask the sites directly and listen for specifics.

Which questions should a reference call cover?

Start with the numbers the site can see for itself, then move to how the program handled trouble. A short script keeps the call comparable across sites and across vendors.

• How many units run today, and how many were installed at go live?
• What share of scheduled runs finished without staff stepping in during the last full quarter?
• How did elevator, door and badge integration behave in the first week?
• Who responded to a fault, and how long did the unit stay out of service?
• Did the program expand, pause or end, and what was the reason?

What do the Moxi 2.0 rollout and ownership change mean?

An August 2026 release reports Moxi 2.0 rolling out to health systems including Endeavor Health Edward Hospital, Providence Saint John's Health Center and Children's Hospital Los Angeles. The release is company reported, so treat each rollout as a claim to confirm at the site.

The company also says it expects to double its hospital footprint each year and to deploy thousands of robots by 2030. That is a projection, and we could not verify it. Ownership is moving as well. Serve Robotics agreed in January 2026 to acquire Diligent Robotics, with closing expected in Q1 2026, and we could not confirm the current status. A buyer should ask which company holds the support contract and whether service terms survive any change in parent.

Where Service Robot Co. fits in a scale review

We start from a different place. Service Robot Co. is an OEM neutral, full service commercial robot integrator. We match the robot to the task across manufacturers, then lease, deploy, integrate, train and service every unit through a nationwide US engineer network, so a hospital deals with one partner rather than a vendor support line plus a separate integrator.

We lease on a three-year term with one fixed lease payment quoted after a site assessment. Hospitals weighing a first delivery robot often start with one leased unit and a clear pilot, then add units once the reference checks hold up. We would not treat any vendor's reach claim, including a count like Moxi's, as settled until named sites confirm it.