Course

When AI Gets It Wrong

A practical field guide to AI's blind spots. You'll see why a chatbot invents facts, why it's brilliant at some things and hopeless at others, why a confident tone means nothing, and how to check its work — so you can lean on it where it earns trust and catch it where it doesn't. No jargon, no code.

6 lessonsBeginner31 min

Start the course Get the app

Course lessons

Why it makes things up

A made-up answer wears the same confident voice as a real one. Here's why — and the tell that gives it away.

›

Teaching it to say 'I don't know'

A model would rather guess than admit a blank. Here's how it learns to recognise the edge of what it knows — and stop there.

›

Vague memory vs fresh notes

Ask from memory and it guesses; hand it the source and it gets exact. The single biggest upgrade to any AI answer.

›

The Swiss-cheese mind

AI ability is spiky, not smooth — genius at some tasks, grade-schooler at others. Knowing the shape is half the battle.

›

Confidence isn't correctness

The sureness you hear is writing style, not a truth meter. A 95%-sure answer can be flat wrong.

›

Check its work

One quick move — source, cross-check, or re-ask — catches most confident mistakes. Build the reflex.

›

Take the full course

Learn it properly — tracked & personalized

Follow the whole course as a guided journey — sequential lessons, progress, streaks and a personalized path, all in the app.

Get the app