📊 How AI Learns from Data

AI learns by looking at huge amounts of examples, called data. For instance, an AI that recognizes cats was shown thousands of cat pictures until it learned the patterns.

The more good, varied examples an AI sees, the better it usually gets at recognizing patterns — but it can still make mistakes, especially with things it hasn't seen much of before.

If the data used to teach an AI is unfair or missing certain groups of people or ideas, the AI can end up unfair too. This is called bias, and it's a real challenge scientists work hard to fix.

Understanding that AI learns from data — not from real understanding — helps us know why it sometimes gets things wrong or treats people unfairly.