Skip to content

AI Safety & Responsible AI

Adversarial example

An input intentionally crafted to cause a model to make a mistake.

Example

A nearly invisible image change causes a classifier to label a stop sign incorrectly.

Why people use it

Understanding “Adversarial example” helps teams identify harms and choose proportionate safeguards.

What you'll hear

“We need to account for Adversarial example before we ship this.”

Related terms