Question:
During a security audit, analysts discover that an attacker repeatedly queried a black-box ML model to infer if specific data points were in the training set. The attacker could determine if an individual’s data was used during training. What threat does this attack represent?
Submit