Will AI replace Physician Assistants?

Some of your tasks will be automated. People who use AI will have the edge.

Task exposure to today's AI

34/100

Moderate exposure

This is the occupation's typical task mix. Your own score can be up to 30 points higher or lower depending on how you work: routine share, people contact, judgment and how much you already use AI.

Get your personal score

Will AI replace Physician Assistants?

Not entirely. Today's AI tools can take on about 34% of the typical work in this occupation, mostly: obtain, compile, and record patient medical data, including health history, progress notes, and results of physical examination. The parts that stay human include: Examine patients to obtain information about their physical condition.; Make tentative diagnoses and decisions about management and treatment of patients.; Instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance..

People who learn to use AI for the automatable parts keep the edge. The 3-phase plan below shows where to start.

Which tasks AI can take on

The largest tasks in this occupation, their share of the working week, and how much of each today's AI tools can do.

TaskAutomatable

Obtain, compile, and record patient medical data, including health history, progress notes, and results of physical examination. 65%

Interpret diagnostic test results for deviations from normal.AI already used 57%

Make tentative diagnoses and decisions about management and treatment of patients. 30%

Instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance. 28%

Prescribe therapy or medication with physician approval. 25%

Examine patients to obtain information about their physical condition. 12%

+ 6 more tasks (23% of the week) in the full test.

Your human edge

Examine patients to obtain information about their physical condition.
Make tentative diagnoses and decisions about management and treatment of patients.
Instruct and counsel patients about prescribed therapeutic regimens, normal growth and development, family planning, emotional problems of daily living, and health maintenance.

AI tools to try

Clinical documentation and transcription assistantsChatGPT / Claude for patient information (no patient data)Clinical decision support in your systems

Related roles

A 3-phase AI learning plan

Phase 1

Less documentation time

AI transcription and note drafting where your organisation allows it.
Patient information leaflets in plain language.
Never put identifiable patient data into public tools.

Project

Use your organisation's approved AI scribe for 20 de-identified consultations, then compare average note-writing minutes per patient against your previous 20 notes.

Phase 2

Evidence and decision support

Guideline and literature summaries with AI, checked in the source.
Understand how decision-support tools reach their output.
Know the limits and liability.

Project

Pick 5 recurring clinical questions from your clinic, have AI summarise the relevant guidelines, verify every claim in the primary source, and log each error found.

Phase 3

AI in care processes

Coordination, scheduling and handover with AI help.
Take part in your organisation's AI projects.
Teach colleagues safe use.

Project

Build an AI-assisted handover and referral template for your team, pilot it for two weeks, and measure minutes saved per shift plus colleague feedback from a short safe-use training session.

Take the free 11-question test