Nib asked, and Nib — askNib's tutor — drew the answer live at a whiteboard. This is the spoken transcript; enable JavaScript to watch it drawn.
How do AI language models like ChatGPT actually work? (for a curious adult who uses chatbots but has never studied computing). Cover, in this order: How does ChatGPT work?; How does it read your words?; How does it learn which words come next?; How does it know which earlier words matter?; Why do AI chatbots hallucinate?; Why does it forget what you said earlier?; How did a text predictor learn to follow instructions?; Does it understand, and how much should you trust it? — page 5: Why Plausible Can Be Wrong
Remember the training game? The model was scored on matching real text. Nobody scored it on whether a sentence was actually true.
When you ask a question, it doesn't open a file. It builds the answer one piece at a time, choosing what fits.
Ask for a source, and the most likely text looks like a citation: author, journal, year. The shape fits, even if the paper doesn't exist.
There's no built-in alarm for 'I don't know.' It always produces a next word, and the tone stays equally confident either way.
Where a fact appeared everywhere, likely and true usually match. Where it's rare, a number, a quote, a name, they drift apart.
So a chatbot isn't lying or broken when it invents a detail; it's doing exactly what it was trained to do. Treat fluent as a style, never as proof.