CyberCertPrepAI Security FundamentalsAdversarial ML & Model AttacksEASYFree question
What is the primary goal of a backdoor (trojan) attack, and how does it differ from an evasion attack?
- A.A backdoor attack degrades all predictions equally; an evasion attack targets only specific classes
- B.A backdoor attack always requires hardware access; an evasion attack is software-only
- C.A backdoor attack targets regression models; an evasion attack targets classifiers
- D.A backdoor attack poisons training data so the model misclassifies inputs containing a specific trigger at inference time; an evasion attack perturbs test inputs at inference time without modifying training data
Why D is correct
Backdoor (trojan) attacks occur during training: poisoned data teaches the model to associate a specific trigger pattern with a target class. At inference, clean inputs behave normally; only inputs with the trigger are misclassified. Evasion attacks craft perturbations at inference time without modifying training. Access level, model type, and uniform degradation do not define this distinction.
Know someone studying for AI Security Fundamentals? Send them this one.