AI in Medicine

A four-week elective for fourth-year medical students

Author

Sean Davis

Published

October 13, 2026

Preface

This is the student-facing book for a four-week, pass/fail elective on AI in medicine for fourth-year medical students. Its one idea: it is your job to learn where AI helps your work, and you learn that by trying. The chapters are a guide, not assignments. They give you a framework; they do not try to teach everything. Each week has one core chapter and a menu to choose from, and most of your learning happens when you use AI to teach yourself. Each chapter stands alone. Nothing in the book needs an install, a paid account, or a real patient.

How to read a chapter

Read what helps you with the week’s work, and skip what does not. Every chapter has the same parts, in the same order:

  1. In short. A box at the top: the main idea, what you do, what you keep, and which sections to read if you are short on time. The subtitle under the chapter title gives the honest total time.
  2. What you’ll learn, and what you need. The objectives, the course competencies the chapter serves, and what you need for the activity. Usually that is only your university account.
  3. An opening scenario. A clinical situation you will recognize, before any jargon.
  4. Why this matters. What changes for you in residency, which starts next July.
  5. How it works. The idea, in prose, with figures and tables.
  6. Watch or listen. One video or podcast episode, with a note on what to listen for and where you can stop.
  7. Do. The activity, written so you can finish it alone. It starts with a task that works, and it ends with something you keep. Wherever it touches the literature, you choose your specialty.
  8. Check. A rubric you score yourself against.
  9. What this changes for you. The standing question, what does this change about how you’ll practice? The scenario from the opening comes back with a better answer.
  10. Go deeper. Papers, talks, and videos. Claims nobody has yet checked are flagged [VERIFY] and left flagged until someone has.

There is no summary at the bottom; the “In short” box at the top does that job. The competencies are named by code, AIM-1 through AIM-9 and AIM-X; the coverage table below says what each one is.

The safety rules are short. Use your university ChatGPT Edu or Copilot account. Never put real patient information into a course activity. Check what matters before you rely on it. You sign the note.

Four ideas recur across the chapters, and each is named the same way wherever it appears. Proxies fail: almost every AI failure in medicine is a stand-in that looked reasonable. Efficiency gains get eaten by volume: per-unit savings are cancelled out by growth in use. The published number is the optimistic one: the most prominently reported statistic is nearly always the best case. Who benefits, who bears the risk, who decides: the people a tool helps are often not the people it puts at risk, or the people who chose it. A chapter that uses one of these says so.

The chapters, week by week

Each week has one core chapter, which goes with Monday’s talk, and a menu to choose from by interest. The syllabus gives the live sessions, the weekly routine, and the due dates. Each week you also write a short learning journal entry about what you tried. Whoever runs the course reads the For instructors appendix.

Week 1, working with AI

Week 2, getting good at it

Week 3, AI and your patients

Week 4, responsibility

Competency coverage

The course tags every chapter to a stable competency ID so it can be mapped to national standards when they arrive. Where the IDs come from, and the frameworks behind them, is in the appendix What the field expects; the coverage rules are in curriculum/competencies.md in the repository. In short: AIM-1 foundations; AIM-2 critical appraisal; AIM-3 bias and equity; AIM-4 ethics, law, and regulation; AIM-5 clinical use; AIM-6 talking to patients; AIM-7 workflow and systems; AIM-8 health data science; AIM-9 research and scholarship; AIM-X professional identity, the closing question.

Table 1: Which material serves which competency. ● the main subject: taught, and practised in the Do or Check. ○ taught, but not practised. Judged from what each chapter teaches and checks (reviewed October 2026), not from the codes the chapter lists.
Material 1 2 3 4 5 6 7 8 9 X
Week 1
Plan your four weeks (Thursday session) ○ ● ○ ○
How these systems work (core) ● ● ○ ○ ○
Week 2
Prompting lab (core) ● ○ ● ○ ○
Critical appraisal ● ● ● ○ ○ ● ○
Bias and equity ○ ● ○ ○ ○
Grounding AI in real sources ● ● ○ ○ ● ○
Week 3
Talking to patients about AI (core) ○ ● ○ ● ○
Agentic coding ○ ● ● ● ○ ○
Week 4
Ethics: what you owe when you sign the note (core) ○ ○ ● ○ ○ ○ ○ ●
Will AI take your job? ● ○ ○ ● ○ ●
What AI costs the planet ● ● ● ○ ○
Course pages and appendices
Patient information and AI tools ●
Learning journal ○ ○ ● ○
Open lab: hear it, draft it, check it ○ ○ ○ ○ ○

The four core chapters make AIM-2, AIM-4, AIM-5, and AIM-6 main subjects, and the week 1 live talks cover AIM-1. AIM-3, bias and equity, is not in a core chapter: it is covered by the bias chapter on the menu. That is a deliberate choice for this offering, made so the core can focus on using AI well. AIM-8, health data science, is not the main subject of any chapter; the foundations chapter covers training and test sets, area under the curve, and dataset shift along the way. Every Check is scored by you against a rubric, so “practised” here means self-assessed, not graded.

How this book was made

This book was built with AI: AI agents drafted most of the text under the author’s direction, and the author decides what is published. The appendix How this book was made and the AI ledger say what the AI did and what a person checked, because the course asks you to disclose and verify, and the book does the same.