How machines sense, reason, and learn — explained from scratch, for everyone.
AI is the science of building systems that do things we'd call smart — recognizing a face, understanding a sentence, or picking a winning move.
THE BIG IDEA
Nobody writes down every rule. The system learns patterns from examples — just like you learn from practice.
Before a computer can spot a cat in a photo or translate a sentence, it has to learn from data. Let's find out what "learning" really means for a machine.
It studies labeled examples — questions paired with the right answers — then answers new questions on its own.
No answer key. It explores the data and finds its own patterns and groupings.
It learns by trying — earning points for good moves and losing them for bad ones.
Information starts at the input, flows through hidden layers that mix and weigh the signals, and arrives at an output — the answer. Practice makes those connections stronger.
Train your very first model to read handwritten numbers.
Teach a system to tell happy reviews from grumpy ones.
Explore where AI gets things wrong — and why that matters.