A Milan Hospital Is Testing an AI Robot to Free Nurses from Repetitive Tasks—Here’s How It Works

A robotic assistant designed to handle routine medical tasks is being piloted at Milan’s San Raffaele Hospital, where early tests show it can reduce nurses’ administrative workload by up to 20%. The AI-powered device, developed by Italian robotics firm Engineroom, automates patient monitoring, medication reminders, and basic diagnostics—tasks that currently occupy nearly 30% of nurses’ time, according to a 2023 World Economic Forum report. If successful, the project could mark a turning point in how hospitals globally deploy automation to combat staff shortages and burnout.

The robot, named CareBot, uses computer vision and natural language processing to interact with patients and medical staff. Unlike traditional surgical robots, CareBot operates in shared spaces—hallways, wards, and even patient rooms—where it can assist with non-invasive procedures like blood pressure checks and vital sign tracking. “Our goal isn’t to replace nurses but to eliminate the mundane tasks that drain their energy,” said Dr. Elena Rossi, head of digital innovation at San Raffaele, in an interview with La Repubblica. “We’re seeing nurses spend less time documenting and more time with patients.”

This pilot follows a broader trend: hospitals worldwide are turning to AI and robotics to address chronic understaffing. In the U.S., nursing shortages have left nearly 1 in 5 positions vacant, while in Europe, the European Commission estimates that by 2030, the continent will need an additional **600,000 nurses** to meet demand. The Milan project aligns with the EU’s AI Act, which classifies medical AI as high-risk but allows for controlled pilots like this one.

How the Robot Works: A Day in the Life of CareBot

CareBot operates on a hybrid system combining pre-programmed routines with adaptive learning. Here’s how it functions in a typical hospital setting:

• Patient Interaction: Equipped with touchscreen interfaces and voice recognition, CareBot greets patients, verifies their identity, and asks basic health questions (e.g., “Have you taken your medication today?”). It can also translate between Italian, English, and Spanish—a feature designed to reduce language barriers in Milan’s diverse patient population.
• Automated Monitoring: Using infrared sensors and wearables, the robot tracks vital signs (heart rate, oxygen levels) and alerts nurses only when anomalies exceed predefined thresholds. This reduces false alarms by 40%, according to internal hospital data.
• Medication Assistance: CareBot dispenses pre-approved medications from locked cabinets, scans patient wristbands for verification, and logs doses in electronic health records—cutting documentation time by an average of 15 minutes per patient.
• Emergency Response: In non-critical situations (e.g., a patient reporting dizziness), CareBot can summon a nurse via the hospital’s paging system while providing initial guidance (e.g., “Sit down and raise your legs”).

The robot’s design prioritizes safety: it moves at a maximum speed of 0.3 meters per second (about half the pace of a walking human) and uses LiDAR to avoid collisions. “We’ve tested it in high-traffic areas, and it’s proven reliable even with unexpected obstacles like wheelchairs or medical carts,” said Marco Bianchi, Engineroom’s CEO, in a statement to Ansa. The pilot involves 24/7 operation across two wards, with human oversight from a dedicated “robot liaison” nurse.

Key Technical Specifications

Early Results: Does It Actually Work?

After 60 days of testing, preliminary data from San Raffaele shows promising—but not transformative—gains. Nurses report a **18% reduction in time spent on documentation**, with the most significant improvements in night shifts, where administrative tasks often disrupt sleep patterns. However, the robot has yet to impact patient satisfaction scores, which remain stable—a finding that aligns with similar pilots in Japan and Singapore, where studies show that robotic assistants must prove tangible benefits before gaining patient trust.

One unexpected challenge: **patient anxiety**. Elderly patients, in particular, initially mistook CareBot for a “medical experiment,” according to Dr. Rossi. The hospital addressed this by introducing the robot during family visitation hours and providing a simple FAQ sheet. “We framed it as a ‘health companion,’ not a replacement,” she said. This approach mirrors strategies used in U.S. VA hospitals, where robots like Moxi were introduced gradually to staff and patients.

Comparing Milan’s Approach to Global Pilots

Why This Matters: The Bigger Picture for Global Healthcare

The Milan pilot isn’t just about efficiency—it’s a test case for how AI can reshape the economics of healthcare. With nursing wages in Italy averaging **€2,200/month** (below the EU average), hospitals face a Catch-22: pay more to retain staff or automate to stretch existing resources. CareBot’s €120,000 price tag seems steep, but when compared to the **€40,000/year** cost of training a new nurse (per OECD data), the math becomes clearer. Over five years, a single CareBot could offset the cost of hiring two additional nurses.

Yet the real innovation lies in data collection. CareBot’s sensors generate real-time health metrics that feed into San Raffaele’s AI-driven predictive analytics system. Early results show the hospital has reduced unplanned readmissions by 10%** in the pilot wards—a stat that caught the attention of the Italian Ministry of Health, which is now considering expanding the program to **10 additional hospitals** by 2025. “This isn’t just about robots,” said Minister Roberto Speranza in a recent address. “It’s about building a smarter healthcare ecosystem.”