technology hard MCQ

Which attack manipulates ML model inputs to cause misclassification while appearing normal to humans?

  1. Data Poisoning
  2. Adversarial Examples
  3. Model Inversion
  4. Membership Inference

Answer: Adversarial Examples

Adversarial examples add imperceptible perturbations to inputs that cause ML models to misclassify. Critical vulnerability for computer vision, NLP systems requiring robust defenses.

Topic AI/ML Security
Exam Relevance Banking, UPSC, SSC
More technology Questions Back to all questions