AI Prompts for Doctors: The Ones That Actually Save You Time

Most "AI prompts for doctors" lists hand you the obvious stuff: summarize a paper, draft a note. Useful, but you already knew AI could do that. What actually changes your week are the handful of prompts that use AI to sharpen your thinking and your delivery, not just your paperwork, so those are the ones I am leading with. Change the details in brackets and they are yours.

One rule before we start, and it matters: never paste a patient's name, MRN, or any identifying detail into an AI tool. Describe the case generically instead ("a 68-year-old with new chest pain"). Treat these tools as a drafting and education aid, never a diagnostic tool or a substitute for your clinical judgment, and verify everything they produce. If you want the fuller version of how to talk to these tools well, that is much of what my book, The Beginner's Guide to the AI Galaxy, is about.

The five most doctors never think to try

Steelman the differential you are anchoring away from. This one quietly makes you a safer clinician, not just a faster one.

I have a [generically described case: 55-year-old with fatigue, mild anemia, and vague abdominal discomfort]. I am leaning toward [a benign explanation]. Steelman the diagnoses I might be anchoring away from. For each, tell me the one finding that would make me take it seriously and the cheapest test to rule it in or out. Push back on my thinking.

Rehearse delivering the hard diagnosis. You do not have to walk into that room cold.

Role-play a patient I am about to tell they have [a serious new diagnosis]. Let me practice breaking the news. React the way a frightened person really would. After a few exchanges, stop and coach me on my word choice, my pacing, and where I was too clinical or too vague. I want to practice empathy before I do this for real.

Translate the plan into language they will actually follow. This is where compliance is won or lost.

Rewrite this diagnosis and treatment plan in plain sixth-grade language a worried patient will actually understand and follow. Keep it accurate, warm, and short. Flag anything a patient commonly misunderstands about it. Plan: [paste it, no identifying details].

Anticipate the questions before you walk in. The ones the patient or family will ask the moment you pause.

A [worried spouse of a patient newly diagnosed with heart failure] is about to ask me questions. List the [8] they are most likely to ask, hardest and most emotional first, and give me a clear, honest, plain-language answer to each so I am not caught flat-footed.

Turn my messy notes into a clean referral letter. Same information, suddenly it reads like you had an hour.

Turn these rough notes into a professional referral letter to a [gastroenterologist]. Lead with the single question I want answered, summarize the relevant history clearly, and keep it concise. Notes: [paste, generically described, no names or MRNs].

The everyday time-savers

These are the staples. They will not surprise you, but they will give you your evenings back.

Draft a patient-education handout.

Write a one-page take-home handout on [managing newly diagnosed type 2 diabetes] for a patient at a [sixth-grade] reading level. Cover what it is, what to do this week, warning signs to call about, and when to follow up. Warm and clear, no jargon.

Write a plain-language after-visit summary.

Turn this visit into a short after-visit summary the patient can read at home: what we found, what we are doing about it, what they need to do, and when to come back. Plain language, reassuring but honest. Visit details: [paste, no identifiers].

Draft the letter you have been putting off.

Draft a [work note / prior-authorization letter / patient letter] about [a patient needing two weeks of modified duty after a back injury]. Professional, specific, and concise. Leave the name and dates blank; I will fill them in.

Explain the labs in language a patient gets.

Explain what these lab results mean in plain language for a patient, one line each, calm and clear, noting which are normal and which we will keep an eye on. Do not give medical advice beyond what the numbers show. Results: [paste values only, no identifiers].

Prep me for boards or recertification.

Quiz me on [acute coronary syndrome management] at a [board] level. Ask one question at a time, wait for my answer, then tell me if I am right and explain the reasoning, especially the distractors. Keep going until I say stop.

Simplify a consult explanation.

A [primary care colleague] is referring me a patient and I need to explain [why I am recommending a particular workup] in two clear sentences a busy clinician will read fast. Draft it, then give me an even shorter version.

How to make any of these yours

These prompts are starting points, not magic words. The single biggest upgrade is telling it exactly who the patient is (generically) and what you actually want, the way you would brief a sharp new resident. The more real context you put in, the better it comes back, which is the whole idea behind talking to AI like a person, not a search box. And if a first answer is close but not right, do not start over, just tell it what to fix, one of the small prompting moves that change everything.

None of this replaces the part only you can do, which is the diagnosis, the judgment, and the responsibility that comes with the white coat. It clears the drafting and the busywork off your desk so you have more of yourself left for the patient in front of you. That is the honest promise of AI for a physician, and it is a good one.