Câu 108: Isaca Advanced in AI Security Management

Which of the following strategies is the MOST effective way to protect against AI data poisoning?

1. A. Ensuring the model is trained on diverse data sources
2. B. Increasing model complexity to better handle data variations
3. C. Using robust data validation techniques and anomaly detection — đáp án hiện tại
4. D. Incorporating more features and data into model training