Triggers when data must be de-identified for analytics, sharing, or ML — k-anonymity, differential privacy, pseudonymization, tokenization, re-identification risk, and synthetic data generation.
