Inference
Inference is the moment a trained model is used: it receives your input and computes an answer. Training happens once (or occasionally); inference happens every time anyone uses the model, so its cost and energy use add up quickly.
In one line, for a 12-year-old
Inference is the AI actually answering you, after all its learning is done.
An example
Each time you press Send in a chatbot, a data centre runs inference to produce the reply, word by word.
Why it matters to people
Where inference runs — on your phone or in a distant data centre — decides where your words travel and which country's laws apply to them.