AI Foundations
Inference
The moment a trained model is used to produce an answer, prediction, or other output.
Example
After training, the spam filter examines a new message and predicts whether it is spam.
Why it matters
It puts an already learned ability to work on each new request.
You'll hear it
“The delay happens while the AI is producing the answer.”
Can you control it?
Set by the team
The people building or running the AI choose this setup. An everyday user generally needs their help to change how this part works.
Common questions
- Can this happen on my own device?
- Yes. Some features run locally, while others send the work to a service elsewhere.
- Is inference the same as training?
- No. Training changes what the system has learned; inference uses that learning to produce a result.
- Why does inference cost money?
- The computers need power and capacity to process each request and create its result.