While it's sometimes okay to interpret the output of a model as a confidence, you do so at your own peril because that's not really what the model is optimizing

While it's sometimes okay to interpret the output of a model as a confidence, you do so at your own peril because that's not really what the model is optimizing. There are often cases where the model outputs a very high "confidence", but the model uncertainty is (or should be) very high. This is generally the case anytime you give the model data that is very far (in some sense) from the data that it's been trained on.