Definition
An adversarial example is an input constructed to make a model produce an incorrect or attacker-chosen result. The input is usually derived from an ordinary example by a perturbation that stays within a defined constraint, such as a small image change, altered audio, or a carefully chosen text variation.
The attack exploits a mismatch between the model's learned decision boundary and the equivalence a defender cares about. Two inputs may look or mean the same to a person while the model maps them to different outputs. In a targeted attack, the attacker wants a particular wrong result. In an untargeted attack, any incorrect result is enough.
Where the term came from
Christian Szegedy and coauthors used adversarial examples in their 2013 paper Intriguing Properties of Neural Networks. They showed that small, deliberate perturbations could cause several image classifiers to misclassify inputs and that some examples transferred between models. Later work expanded the term across modalities and threat models.
Distinguish it from nearby terms
A naturally difficult example is not necessarily adversarial because no attacker constructed it. Data poisoning changes training data; an adversarial example usually attacks inference. Prompt injection uses content to redirect instruction following or tool behavior. Text attacks can blur these boundaries, so state the attack mechanism rather than relying on the label alone.
Check your understanding
A stop sign image is altered so a classifier reads it as a speed-limit sign. What must the evaluation report? The attacker's target, allowed perturbation, physical or digital setting, human perceptibility, model access, and whether the attack transfers to the deployed pipeline.