Câu 106: Isaca Advanced in AI Security Management

Implementing which of the following would MOST effectively address bias in generative AI models?

1. A. Data minimization
2. B. Data augmentation
3. C. Adversarial training
4. D. Fairness constraints — đáp án hiện tại