AI Use Cases for Doctors

← Browse workflows

Teaching & learning

Summarize resident evaluations before the competency committee

I have a year of end-of-rotation comments on every resident and no time to pull them together before the Clinical Competency Committee meets.

Small retrospective evaluation De-identified text only Organization involvement

Source check: 2026-11-11 · Physician review: not yet recorded

The workflow

  1. Obtain program approval for a workspace that covers confidential employee/trainee evaluations.
  2. Keep source assessments and stable IDs so every compiled statement can be traced back.
  3. Use the tool to organize comments by resident, without inventing judgments or discarding negative comments.
  4. Have faculty compare every summary with the original assessments before committee use.

Tools and cost

USD unless stated otherwise. Billing periods and eligibility matter. A tool listed here was not necessarily tested in the cited study.

ChatGPT
$0 · Free plan; optional paid plans · Price source
OpenAI's general-purpose AI chat assistant (the regular version, not ChatGPT for Clinicians).

Individual self-service
Consumer Free is separate from the BAA-eligible clinician workspace.

Microsoft Copilot
$0 · consumer Copilot; Microsoft 365 plans separate · Price source
Microsoft's general-purpose AI chat assistant, with a free version.

Individual access
Do not infer commercial Microsoft 365 coverage for consumer Copilot.

What the evidence supports

In one residency’s 277 assessments for 25 residents, AI labels were correct for 59.7% of comments. Recall was 67.8% for strengths but only 23.9% for areas needing improvement.

Primary study · Read the source ↗

Where it goes wrong

The tool missed most improvement comments. It should not judge competency or replace reading the source assessments. Removing names or disabling model training does not establish approval to upload confidential evaluations.

← All use cases