3.1 How AI Works
AI takes in data, finds patterns during training, then uses those patterns to make predictions on new stuff.
Chapter 3: Fundamentals of AI
Every AI system, from a spam filter to a chatbot, follows the same basic loop: input → process → output, with learning happening behind the scenes.
During training, the AI is shown loads of examples (say, 100,000 emails labelled "spam" or "not spam"). It adjusts its internal settings until it gets good at telling them apart. That trained result is called a model.
Then comes inference - the AI using what it learned on brand-new data. A fresh email arrives, the model checks it against the patterns it learned, and outputs a prediction: spam or not. If it gets feedback ("this wasn't spam!"), that can be used to improve it later.
Key takeaways
Training = learning patterns from examples
Model = the trained result
Inference = using the model on new data