AWS Certified AI Practitioner AIF-C01 Practice Question

Inferencing (also called prediction or scoring) is the phase that begins after training is complete. During inference, a previously trained model is executed on new, unseen data to generate outputs such as classifications, probabilities, or recommended actions. The model's learned parameters remain fixed; no additional weight updates or hyperparameter tuning occur. Activities like updating weights with labeled data, performing hyperparameter searches, or initializing parameters belong to the training workflow, not to inferencing.