AI tools can state wrong facts in a confident tone, including names, figures and sources that do not exist. This is a known behaviour, not a rare fault.
The tool produces text that fits the pattern of the question. It does not look facts up unless it is connected to a source, so a plausible answer can still be wrong.
Specific details such as dates, quotations, references and numbers. The more precise the claim, the more it deserves a check.
Check anything you will rely on against a source you trust. Ask the AI to say where it is unsure, and treat the answer as a draft.
Give it the source material in the prompt and ask it to answer only from that text. This does not remove the risk, but it narrows it.
Our free page 'How to prompt better' has a copyable template for the four questions: goal, background, limits and done when. Open How to prompt better
Last reviewed: 2026-11-02
#AI #Prompting #ChatGPT #Claude #Productivity
This page gives general information and is not legal, financial or medical advice. [email protected]