Which control MOST effectively prevents an AI model from being retrained on data that has been contaminated by a data poisoning attack?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — b is correct because anomaly detection on incoming training data — flagging statistical outliers, unusual label distributions, or anomalous feature patterns — is the most direct preventive control against data poisoning by detecting contaminated data before it enters the retraining pipeline. A protects data at rest from unauthorized access but does not detect poisoning.
Full explanation below image
Full Explanation
B is correct because anomaly detection on incoming training data — flagging statistical outliers, unusual label distributions, or anomalous feature patterns — is the most direct preventive control against data poisoning by detecting contaminated data before it enters the retraining pipeline. A protects data at rest from unauthorized access but does not detect poisoning. C prevents internet-based attacks but not insider or supply-chain poisoning. D reduces model capacity but does not prevent poisoning from entering the training set.