Since its launch in July 2025, MedBot has saved 28 man-hours a month and about $15,400 a year at National University Hospital (NUH) in Singapore. Close to 1,300 patients were surveyed over eight months: 98.7% found MedBot easy to use, 98.9% agreed its information was clear and easy to understand, and 96% felt comfortable proceeding with their medications after using it.
"Every piece of AI-generated counselling content goes through a human-in-the-loop process — our pharmacists review and validate all content before it enters the database. This was non-negotiable," said Jack Lim, Data Scientist at the Kent Ridge Office of Innovation, NUH, who led the technical build.
Pharmacists at NUH were spending three to 20 minutes per patient on medication counselling. With some 2,500 prescriptions dispensed daily across NUH's outpatient and satellite pharmacies, these sessions were adding to waiting times and operational bottlenecks and tying up clinical expertise in standardised information.
MedBot is a generative AI-powered virtual pharmacy assistant that gives patients medication information before they collect their prescriptions. It is built on the AWS Healthcare Commercial Cloud and integrated with the hospital's electronic medical records and the medication database on H-Cloud Data Centre. Pharmacy typists identify suitable patients when taking medication orders, brief them on MedBot and direct them to dedicated kiosks at the dispensing counter, currently at the Kent Ridge Wing Pharmacy and the satellite pharmacy at the National University Centre for Oral Health, Singapore. The system delivers from a pre-approved database rather than live AI generation, and pharmacists review all content before it enters that database. Safety guardrails include allergy acknowledgement checks before use, pharmacist review panels and clear escalation pathways.
The team started with 10 carefully selected medications and expanded to 66 only after strict safety milestones were met. A multidisciplinary team drove the work, with Principal Clinical Pharmacist Goh Sok Hiang overseeing clinical strategy and senior pharmacist Wong Hong Yun steering the patient-facing experience. MedBot was one of the first NUH projects to move patient medication data in real time between the on-premise data centre and the cloud environment.
NUH plans to scale MedBot to other pharmacies, starting with those that frequently dispense simple and low-risk medications. Enhancements under way include an interactive chatbot that lets patients ask MedBot questions about their medications, and pharmacists have suggested a mobile version. Lim said the team is looking at moving to a human-over-the-loop concept, where AI operates with greater autonomy while humans keep oversight and control.