How AI Counts Words to Pick the Next One

When AI writes a sentence, it picks one word at a time. For each spot, it guesses what word usually comes next based on tons of writing it has read. It's like a super fast game of fill-in-the-blank.

Some examples

• After 'Once upon a' the AI guesses 'time' because that's what usually comes next.
• After 'peanut butter and' the AI guesses 'jelly' because most sentences go that way.
• If you start with 'The cat sat on the' the AI might pick 'mat' or 'rug'.
• AI doesn't know what the words mean — it just knows which ones tend to go together.

Try it!

Open a chatbot and start a sentence like 'My favorite snack is'. Don't finish it. See what the AI picks. Try a few times — does it pick the same word every time?

Practice this safely

Try this with a low-stakes example and a trusted adult nearby. The goal is to notice how AI talks about next-word prediction, not to let it make the decision for you.

1. 1Ask AI to explain next-word prediction in plain language, then underline anything that sounds uncertain or too broad.
2. 2Give it one detail from "How AI Counts Words to Pick the Next One" and ask for two possible next steps plus one reason each step might be wrong.
3. 3Check probability against a trusted source, teacher, adult, expert, or original document before you use it.