Which is a key objective of secure ML lifecycle?

• A. Data minimization and consent
• B. Increasing model size
• C. Reducing latency at all costs
• D. Ignoring governance

Answer: (A)

Prioritizing data minimization and consent helps protect privacy and ensure lawful, accountable use of data throughout the machine learning workflow. By collecting only what is truly necessary for the model’s purpose and obtaining clear consent from individuals, you reduce the exposure of sensitive information, limit retention, and strengthen governance and rights management. This foundation supports secure data handling, access controls, and compliance with privacy regulations, which are essential for trustworthy ML.

The other options miss this security-and-governance focus. Merely increasing model size doesn’t address privacy or data governance. Chasing the lowest latency at all costs can pressure security and privacy decisions. Ignoring governance eliminates the policies and oversight that keep data handling responsible and compliant.