Ambient scribe for 2,000 clinicians
The model works; the clinicians don't trust it yet.
Open the live lab · preloaded to this scenario
Adoption Readiness Decision Instrument
Context
A health system pilots an ambient AI scribe for 2,000 clinicians. Note quality is good in the demo, but physicians distrust auto generated notes they're legally accountable for, and nothing in the comp model rewards using it.
The decision
Gate on trust and workflow fit, not accuracy. At a composite in the high-50s this is a Hold, scaling now burns clinician goodwill you can't re buy.
What most miss
Everyone optimizes the model's word error rate; adoption dies on 'I'm liable for this note and I didn't write it.' Trust and a clean override are the real ramp.
Stakes
A failed clinical rollout doesn't just waste spend, it poisons the next three AI initiatives with the medical staff.
Studied · Operating Model and Transformation Leadership Artifacts · verified 2026-07-03
Sources: Ambient clinical-documentation adoption patterns; Clinician trust / note-liability and override literature