In smaller communities, it’s common for patients to cycle through multiple touchpoints—provider visits, lab processing, imaging reads, referrals, and then re-checks. That structure can create predictable failure points:
- Abnormal results not flagged quickly enough or not communicated clearly.
- Follow-up instructions that weren’t practical to carry out (or weren’t tracked).
- Hand-offs between providers where symptoms and test results didn’t fully carry over.
- Diagnostic uncertainty that wasn’t resolved after repeat visits.
When AI tools were part of the workflow, the risk often isn’t that software is “evil”—it’s that the system’s suggestion can become the path of least resistance. If the team treated an automated output as a conclusion rather than a prompt to verify, the legal question becomes whether that approach met the standard of care in North Carolina.


