A security team is tasked with implementing data poisoning defenses for an AI model used in fraud detection that is retrained weekly on new transaction data. Which control most directly prevents poisoned data from corrupting the model during retraining?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — b is correct because data poisoning attacks work by injecting malicious samples into training data that cause the model to learn incorrect behaviors (e.g., approving fraudulent transactions). The most direct defense is to inspect and validate new training data before it enters the retraining pipeline.
Full explanation below image
Full Explanation
B is correct because data poisoning attacks work by injecting malicious samples into training data that cause the model to learn incorrect behaviors (e.g., approving fraudulent transactions). The most direct defense is to inspect and validate new training data before it enters the retraining pipeline. Statistical anomaly detection can flag unusual data distribution shifts (e.g., sudden spike in a particular transaction pattern), and human review of flagged samples catches adversarially crafted records before they corrupt the model. A is wrong because DDoS protection defends against network-layer volumetric attacks; it does not prevent semantic data poisoning through legitimate-looking data samples. C is wrong because encrypting training data protects its confidentiality at rest but does not prevent a legitimate data source from being compromised and serving poisoned records. D is wrong because differential privacy adds noise to model outputs to protect training data privacy; it does not prevent data poisoning attacks on the training inputs.