A Monday at a Dubai Clinic: AI Handled 14 Patient Inquiries
Jas AI Blog ·
Example scenario — all characters, names, and interactions in this post are fictional. The workflow and channel behaviors described reflect how Jas AI operates.
Dr. Layla's aesthetic clinic in Jumeirah opens at 9am. But the first message arrives at 8:05.
Here's how a Monday unfolded — and what the AI handled, and what it didn't.
08:05 — The Promotion Question
A WhatsApp message: "Is laser hair removal still running the promotion from last month?"
The AI checks the knowledge base. No active promotion is listed. It replies: "Our current pricing starts at AED 350 per session. Would you like to book a consultation to discuss a package?"
The patient asks about bundles. The AI confirms pricing structure and logs the inquiry in the CRM.
09:20 — Three Inquiries, Three Languages
Within 20 minutes, three messages arrive.
Patient A (Arabic, WhatsApp): Post-care instructions for a recent Botox treatment. Full response in Arabic, under 60 seconds.
Patient B (English, Instagram DM): Weekend hours. Answered immediately.
Patient C (Arabic, web chat): "Are the results of the skin-lightening treatment permanent?" The AI replies: "Results depend on your skin type and the treatment plan our doctor recommends. We'd suggest a short consultation — shall I check the next available slot?" It does not speculate about outcomes.
10:40 — A Returning Patient
Sara messages via Instagram: "Hi, I'm back in Dubai. I'd like to book my next session."
Her CRM profile shows the channel she used previously, the treatment she asked about, and a coordinator note. The AI greets her by name and confirms Thursday at 2pm. The reservation module logs the booking.
11:15 — The Message the AI Doesn't Answer Directly
"I had a reaction after my last treatment here. I'm very upset."
The AI acknowledges immediately and connects the patient to the coordinator. Live agent mode activates. The coordinator takes over within four minutes. This is the line AI doesn't cross: complaints requiring clinical judgment get instant escalation.
12:30 — The Afternoon FAQ Window
Six routine inquiries arrive across three channels: parking validation, product availability, gift voucher, cancellation policy, directions, age requirement for a treatment. All six answered by AI. None escalated.
15:30 — The Consent Question
"My daughter is 16. Can she have filler treatment without me being there?"
The AI flags this for the coordinator without attempting to answer. The coordinator calls back and explains the parental consent process.
End of Day
| Category | Count |
|---|---|
| Routine FAQ responses handled by AI | 9 |
| Booking or availability confirmations | 2 |
| Escalations to the human team | 3 |
| Total inquiries | 14 |
What AI Doesn't Do in a Clinical Setting
It doesn't diagnose or recommend treatments. It doesn't speculate on outcomes. It doesn't override escalation rules for messages that touch medical judgments. A correct "I'll connect you with the team" is always more valuable than a confident wrong answer.
FAQ
Can AI reply in Arabic and English in the same conversation? Yes. Jas AI supports Arabic, English, Turkish, and German natively.
What happens when AI doesn't know the answer? It says so and escalates. The knowledge base defines what AI can answer — everything else triggers a handover.
Does the AI handle Instagram DMs and WhatsApp separately? All channels feed into the same system. A patient recognized on WhatsApp is the same contact via Instagram — their CRM profile links both interactions.
How quickly can a clinic set this up? The first knowledge base and channel connections typically come together in 2–3 days.
Explore Jas AI for clinics: jasnow.ai