How Artificial Intelligence Can Be Tricked And Fooled By 'Adversarial Examples' Algorithms can be so sophisticated that they can exceed humans in tasks. However, unlike humans, they are prone to a type of problem called an "adversarial examples."
