What is machine learning?
The computer learns patterns from examples instead of a programmer writing every rule.
What is it?
In machine learning a computer learns patterns from examples instead of a programmer writing every rule. From many examples it builds a model that then assesses new cases. Most of today’s AI is based on machine learning.
An everyday example
A model trained on examples of email predicts whether a new message is spam.
Why does it matter?
Machine learning solves tasks for which rules would be impossible to write by hand, such as recognising faces or speech.
A common misconception
That the model understands causes. It learns associations in the data, which is not the same as understanding why something happens.
In more detail
How does it start?
We define the task, prepare the data, and choose a model and a measure of success.
What does “error” mean?
The difference between the desired output and the actual one. During training, the process tries to shrink that difference.
Does a learned pattern mean a cause?
No. If two things often appear together, that does not mean one causes the other.
Where is machine learning used?
In forecasts, fraud detection, image classification, recommendations and many other tasks.
Remember: Learning a pattern is not the same as understanding a cause.
Reliable sources
Last reviewed 1 November 2026. An explainer, not professional, legal or medical advice. For rules and figures that change, check the official version.