Blog
Notes on safe AI in healthcare
How we think about guardrails, intake, and keeping clinical judgment where it belongs.
SafetyMay 10, 2026 5 min
Why triage rules belong in code, not prompts
Burying safety logic in a system prompt is a liability. Here's how CareGuard makes every urgency decision auditable.
Read moreProductApril 28, 2026 4 min
Structured intake that clinicians actually trust
Free-text chat is easy. Turning it into a clean, typed record your EHR can ingest is the hard part.
Read moreSafetyApril 15, 2026 3 min
Decision-support, not diagnosis: drawing the line
Where an AI health assistant should stop — and how CareGuard enforces that boundary everywhere.
Read more