How to explain artificial intelligence to a child

Artificial intelligence is computer software that learns patterns from examples and uses them to make predictions or decisions. To explain it to a child, skip the technical machinery and start with something they already know: AI is like a student who gets better at a task by seeing thousands of examples. Show them a photo app that recognises faces, or a voice assistant that answers questions, and explain that the computer learned to do this by studying huge numbers of pictures or sentences — not by being given a fixed rulebook for every case.
The clearest way in is the idea of learning from examples. Imagine showing a child a thousand pictures of cats and a thousand of dogs, then asking them to sort a new picture. After enough examples, they spot the pattern — pointy ears, whiskers, tail shape. AI works the same way: it is shown labelled examples until it can guess correctly on things it has never seen before. This "learning from examples" framing is accurate, avoids scary or magical language, and gives a child a mental model they can build on later.
It also helps to be honest about what AI cannot do. AI does not understand the world the way a person does; it finds patterns, and it can be confidently wrong. A useful example: an AI trained only on pictures of cats indoors might get confused by a cat outdoors, because it never saw that pattern. Teaching children that AI makes mistakes — and that a human should check important decisions — builds healthy skepticism early. This matters more every year, as children grow up surrounded by AI-generated text, images, and recommendations.
Explaining AI well to a child is really about building three habits: curiosity about how everyday technology works, comfort with the idea that machines learn from data, and caution about trusting any single answer. A child who understands that AI is a pattern-finder, not a magic oracle, is better prepared for a world where these tools are everywhere. The goal is not to turn a 10-year-old into a programmer, but to give them an accurate, jargon-free mental model they can carry into a future where AI is simply part of daily life.