IAPP Artificial Intelligence Governance Professional (AIGP) — Question 13

What is the best method to apply to an AI model so that it is statistically unlikely that a specific piece of training data can be identified from a model’s output?

Answer options

• A. Homomorphic encryption.
• B. Data sharding.
• C. Differential privacy.
• D. Data compartmentalization.

Correct answer: C

Explanation

Differential privacy is the correct method as it adds randomness to the model's output, making it difficult to trace back to specific data points. Homomorphic encryption protects data in transit but does not address output identification. Data sharding and data compartmentalization focus on data organization rather than ensuring privacy in model outputs.