Free lesson preview · Lesson 6 of 10 · 11 min

Spotting When AI Is Wrong

Confidently catch AI mistakes and hallucinations, and know exactly what to verify before you rely on it.

Here's the skill that keeps you safe: knowing that a confident answer is not the same as a correct answer. The AI writes everything in the same self-assured tone whether it's certain or completely making it up. That invented-but-confident stuff is called a hallucination — a made-up fact, statistic, quote, price, law, or source that sounds real.

The good news: you don't need to verify everything. You need to verify the things where being wrong actually costs you. Split every answer in your head into two buckets: - Trust the wording — tone, phrasing, structure, ideas, drafts. If it words your email nicely, that's just language; there's no 'fact' to be wrong about. - Verify the facts — any number, price, date, name, statistic, legal or tax claim, medical claim, or 'according to [source]'. Check these against your own records or a reliable source before you send or act.