What is Data Poisoning?

An attack where an adversary intentionally inserts, alters, or mislabels data in a machine-learning model’s training dataset to manipulate its behavior or performance. This deliberate interference is distinct from accidental errors or unintentional bias. For instance, an attacker might add mislabeled images to a facial-recognition dataset, causing misidentification or failure to recognize individuals. These attacks can result in biased decisions, security vulnerabilities, or unreliable performance.