Seol mar théacs é seo: Adversarial attacks can deceive AI systems, leading to misclassification or incorrect decisions