How it “thinks”
What’s behind the answers, and why it sometimes makes things up.
Free course · about 1 hour
ChatGPT, Claude, Gemini, Mistral, Copilot… They all work in a similar way. In 8 short lessons you’ll learn to ask well, to know when to trust the answer and to protect your data. No jargon, lots of examples to try.
Every day, millions of people ask an AI assistant for help: to write, to get organised, to understand.
Behind the scenes, the AI picks the most likely word, one after another. It doesn’t look things up in a store of truth.
That’s why how you ask matters: the more context you give (who you are, what you need, tone, format), the more useful the answer.
It can also be wrong with great confidence. Numbers, dates and sources always need checking.
And some things never belong in a chat: passwords, bank details, health information.
In 8 short lessons you’ll learn to do all of this. Shall we start?
What’s behind the answers, and why it sometimes makes things up.
Clear requests = useful answers. With ready-made templates to copy.
The questions to ask yourself before trusting what you read.
What never to type, and when to say you used AI.