If the protection layer goes down
Not a single character goes to the cloud AI. Sanad continues with limited canned replies until the protection layer is back.
The first question every clinic manager asks: where do patient messages go? Here is a message's journey through Sanad, step by step.
Not a single character goes to the cloud AI. Sanad continues with limited canned replies until the protection layer is back.
Each clinic has its own database, and one clinic's messages never mix with another's. An answer about one specific patient never becomes a general answer for others.
Available as an option: we run the AI from Google Cloud's Dammam region, and Sanad refuses to run on any model we haven't confirmed processes inside the Kingdom.
Only your clinic's staff, each with their own account and PIN. Sensitive actions, like cancelling a doctor's day, are for supervisors only, and every action is logged with who did it and when.
Reminders only go to patients who agreed, and marketing messages need separate consent. A patient can stop everything by replying «stop».
Sanad is a receptionist, not a doctor. If a patient describes serious symptoms, it stops the booking, directs them to 997 and alerts reception.
Sanad was designed with Saudi Arabia's Personal Data Protection Law (PDPL), issued by SDAIA, in mind. We don't claim an official compliance certificate, because that comes from an accredited body after an audit, not from us.
Automatic name removal greatly reduces the data that leaves your clinic, but we don't claim it is 100% perfect. That's why the local protection layer reads every message before anything else.
Have a data question we didn't answer here? Contact us, and we'll send you the full technical details.