Evaluation & Quality
Uncertainty quantification
Estimating how uncertain an AI's answer or prediction is.
Example
A forecast includes a range of plausible outcomes rather than one number.
Why people use it
It helps people judge how much confidence to place in an estimate.
What you'll hear
“How wide is the range of plausible results?”
What this means for you
Ask what kind of uncertainty is measured and how the estimate was tested.
Can you control it?
No
No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.
Common questions
- Can uncertainty estimates be wrong too?
- Yes. They depend on the information and assumptions used, so unfamiliar situations can make them misleading.
- Is a narrow range always better?
- No. A range that is too narrow can give a false impression of certainty.
- Can uncertainty have several causes?
- Yes. It may come from noisy measurements, missing examples or assumptions that do not fit the situation.