Question
Select the answer that correctly completes the sentence.
Explanation
When a generative AI model produces output that seems realistic but contains incorrect information, the behavior is known as model inaccuracy, often called fabrication or hallucination. Data leakage is the exposure of sensitive information. Overreliance describes people trusting output without checking it, so it is a user behavior rather than a model behavior. Prompt injection is an attack in which malicious instructions in the input try to make the model act in unintended ways.
