Most health systems
deploying clinical AI
have named one seat.
The governance seat.
A CMO.
A CMIO.
A General Counsel.
Someone at the top
who approved the deployment
and signed the charter.
That seat is named.
The other seat is not.
The second seat
sits at the point of care.
It belongs to the person
who receives the recommendation
and makes the call.
On a specific patient.
At a specific moment.
The attending.
The Chief of Service.
The treating clinician.
Two seats.
One name.
The gap is open.
The Governance Owner
holds authority over the system.
↳ Approved this AI for use.
↳ Can run it or pull it.
↳ Owns the harm at the system level.
The Decision Owner
holds the call at the point of care.
↳ Received the recommendation.
↳ Documented the decision and why.
↳ Defends it when the regulator asks.
Same deployment.
Same charter.
Two different seats.
Two seats. Two names. One gap to close.
The regulator is not reading your charter
to understand your philosophy.
They are reading it
to find the two names.
When they walk in,
you already know
whether your institution can produce both.
Mo Johnson, MD MBA is a cardiothoracic surgeon and the founder of GPe Research. Field Notes are short dispatches from the clinical AI accountability frontier, published alongside the MedicoVigilance™ newsletter at medicovigilance.org.
Follow the work on LinkedIn: linkedin.com/in/mo-johnson