What's in this video
- Build an AI symptom checker from a healthcare workflow.
- Add guardrails that inspect requests for listed personal data.
- Watch the system stop a request before patient data reaches the model.
Frequently asked questions
Topics from this video and why governed AI matters.
Patients describe how they feel and add age and gender; in seconds they get evidence-based guidance (emergency, urgent, routine, or self-care). Sentinel replaces email addresses, phone numbers, Social Security numbers, credit card numbers, IBANs, IP and MAC addresses, and API keys, cloud credentials, and tokens in known formats. Bank account numbers, driver's licenses, tax IDs and EINs, national IDs, VINs, dates of birth, and medical record numbers are replaced when a label sits next to them, such as "Account number:" or "MRN:". For requests routed through Sentinel, the audit log records request metadata.
HIPAA and trust require that PHI never reaches the LLM. Best practice is to detect and block PII at the prompt level before data leaves your control. SUPERWISE guardrails strip or block identifiable data before inference, and every decision is recorded, giving health systems a defensible audit trail.
Every assessment is saved with full context, timestamps, and red flags highlighted. Analytics cover distribution, trends, and export for compliance and care improvement. Governed AI needs this visibility so clinicians and compliance teams see what the system did and why.
Patients and clinicians need to understand why a triage decision was made. SUPERWISE keeps every decision auditable and blocks PII by default. That combination is what makes AI safe for telehealth and population health.




