Clinical AI Literacy: From Prompt to Governed Workflow
Learn to specify, inspect, compare, test, and govern the systems that place large language models inside clinical workflows.
Applied Project / idea
Specify and test one bounded clinical AI workflow with approved evidence, narrow permissions, deterministic checks, a physician checkpoint, and an audit trail.
The Brief
Choose one recurring clinical or clinical-adjacent workflow. Define its evidence boundary, permitted tools, verification rules, physician checkpoint, execution limit, and audit record before granting the system any consequential authority.
The finished project should include a SCOPE specification, Harness Card, evidence map, deterministic verification plan, three-error check, physician checkpoint, and five synthetic or properly de-identified test cases.
Skills Under Load
Preparation
The lessons define the vocabulary and constraints. The project tests whether you can use them together.
Course Context