Four rules apply on every channel, with the exact lines patients hear. Inside the inbox, Inbox Copilot adds clinical guardrails for staff. And the limitations are stated plainly.
Hello, you have reached [Clinic]. I am an AI assistant…
Example scripts. Each clinic's wording is validated with its team before go-live; the rules themselves do not change.
Copilot works for staff, never for patients. When a message contains red-flag symptoms, it raises an Urgent Clinical Alert, blocks a generic administrative reply and starts the hard-coded escalation workflow: the charge nurse is notified and 998 guidance is prepared.
My father has chest pain and his left arm feels numb. Can we come in tomorrow?
21:47 · patient
Thanks for your message! Our next available slot is tomorrow at 10:00.
Generic reply blocked
Flows are validated with each clinic before go-live, and every conversation is logged for review.
Patients can know they are dealing with an AI, ask for a person, and request access, correction, deletion, restriction or objection. Requests go through their healthcare provider, which controls the data; VoclyAI supports the provider in responding.
Asks for access or deletion
Receives and decides
Supports the response
Seven nights on your after-hours line. No IT disruption, and an ROI audit at the end.
# 7-day pilot period period: 7 days pilot: After-Hours # voice: nights, weekends, holidays timeline: day 0 setup no IT project day 1-7 live AI disclosed on every call day 8 roi audit answered · captured · recovered rules: disclosure, human on request, 998, no medical advice