Adversarial examples, poisoning, model extraction and membership inference. What a model memorises from its training data and how an attacker draws it back out. The provenance marks that identify synthetic content.
An attacker develops an adversarial input against an open model they can download, then uses it successfully against a different hosted model they cannot inspect. What property does this demonstrate?
AExtraction, since the attacker has recovered the hosted model.
BTransferability, since attacks built against one model often work against another trained for the same task.
CPoisoning, since the attacker influenced the hosted model's training.
DMembership inference, since the attacker has learned about the training data.