All modules
Module 01 · 10 minBeginner

What is artificial intelligence?

A plain-language tour for clinicians: how AI differs from regular software, what machine learning actually means, a high-level view of neural networks, and how generative AI writes the note in your EHR.

A working definition

Artificial intelligence is software that performs tasks we usually associate with human thinking — recognizing images, understanding language, making predictions — by learning from data rather than from hand-coded rules.

Think of a medical resident: nobody hands them a rulebook for every possible patient. They learn by seeing many cases, getting feedback, and refining their judgment. AI works in a similar way — minus the lived experience, ethics, and accountability.

Traditional software vs. AI

Input
Patient data
Hand-written rules
if temp > 38.0 → "fever"
Output
Same answer, every time
Traditional software

A developer writes explicit rules. The behavior is predictable, transparent, and only as smart as the rules.

AI software

The model learns patterns from data. It can handle messy inputs the developer never anticipated — but it can also be wrong in surprising ways.

Interactive activity · 3–5 min

Teach the Computer: It's Harder Than It Looks

Machine learning finds patterns in the examples and labels people give it.

Round 1 — Easy training. Let's teach our computer the difference between fruits and vegetables. Drag each card into a bin (or tap the card, then tap a bin).
Training examples: 0 / 80%
Fruit
Drop fruits here
Vegetable
Drop vegetables here

How a Neural Network Recognizes Patterns

Explore each part of the network. Click a tab to see how that piece works.

Input LayerHidden Layer 1Hidden Layer 2Output LayerViral illnessBacterial illness

The Input Layer receives information from the patient or healthcare record.

  • Age
  • Symptoms
  • Vital signs
  • Laboratory results
  • Medical history

What makes generative AI different

Traditional ML classifies (is this pneumonia?). Generative AI creates — new text, images, summaries, or code that didn't exist before.

Predictive AI

Picks an answer from existing options. "Is this lesion benign or malignant?"

Generative AI

Produces something new. "Draft a patient-friendly summary of this visit." It predicts the next most-likely word, over and over, until a paragraph appears.

Important: Because generative AI predicts likely words — not true ones — it can sound confident while being wrong. Always verify clinical content.

Clinical examples

Where you've likely already seen these ideas in practice.

Ambient documentation

An AI scribe listens to the visit and drafts a note. You review, edit, and sign.

Imaging triage

An ML model flags chest X-rays that may show pneumothorax, moving them up the radiologist's queue.

Patient-friendly summaries

Generative AI rewrites a discharge plan in 6th-grade reading level for your patient to take home.

Key takeaways
  • AI is software that learns patterns from data — not from hand-written rules.
  • Machine learning improves with examples; quality of data drives quality of output.
  • Neural networks are layered math, loosely inspired by the brain — useful, not magical.
  • Generative AI creates new content by predicting the next likely word, pixel, or token.
  • AI is a tool that assists clinicians. The human stays accountable for every decision.

Concept cards

Flip each card to reveal a plain-language definition and clinical example.

Quick check

A colleague says 'machine learning is just programming with extra steps.' What's the most accurate response?

Check your understanding

Five quick questions on the core ideas from this module.

Knowledge check
Question 1 of 5

You're not signed in — this score won't be saved or counted toward your level average. Sign in first to record it.

What's the biggest difference between traditional software and AI?

Score: 0
All modules