Manipulating input data to deceive machine learning models, causing them to make incorrect predictions or classifications.
Generality: 650