Neural Network
A way of building AI loosely inspired by how brain cells connect and pass signals to one another. It is made of many small, simple units linked in layers, and it learns by gently adjusting the strength of those links as it sees examples. This design is called a neural network, and it powers most modern AI.
A neural network is built from lots of tiny units that each do something very simple: they take in numbers, combine them, and pass a result along to the next unit. On their own these units are not clever. Connected together in layers, and tuned through training, they can do remarkable things like understanding a sentence or recognising a face.
The name comes from a loose comparison with the brain, where neurons pass signals between one another. It is worth being honest that the comparison is rough. A neural network is not a digital brain, and it does not think the way we do. It is a mathematical structure that happens to borrow the shape of the idea.
Learning happens through gentle adjustment. Each connection has a strength, and during training those strengths are nudged up or down so the network’s answers gradually improve. This is the machine learning process at work, repeated across an enormous number of examples.
You do not need to picture the maths to use AI well. What is worth knowing is that when a chat assistant answers you, a large neural network is doing the work behind the scenes. The large language models behind today’s chat tools are neural networks with a truly vast number of these connections.