Exam AWS Certified AI Practitioner AIF-C01 topic 1 question 117 discussion

Explainability helps with understanding the cause of a prediction, auditing, and meeting regulatory requirements. Explainability is part of Operational excellence pillar best practices which rolls up to the Business goal identification lifecycle phase of the Well-Architected machine learning design principles. https://docs.aws.amazon.com/wellarchitected/latest/machine-learning-lens/mloe-02.html

The business goal identification phase is crucial for determining compliance and regulatory requirements because it establishes the scope of the model’s application, including legal constraints, privacy regulations (like GDPR or HIPAA), and ethical considerations. These requirements are often aligned with the business objectives at the start of the project to ensure the solution remains compliant.

A clear problem definition keeps the entire ML team aligned on what success looks like. However, this step is far from straightforward. For example, setting appropriate risk thresholds for fraud detection involves balancing regulatory requirements (like GDPR, AML, and KYC) with business priorities and operational constraints.

c. Data collection

The correct answer is D. Business goal identification phase establishes all requirements including compliance and regulatory.

C. Data collection Explanation: The data collection phase of the ML lifecycle is where compliance and regulatory requirements are primarily determined. During this phase, it's important to ensure that the data being gathered complies with legal and regulatory standards, such as data privacy laws (e.g., GDPR, HIPAA). Compliance considerations include ensuring that data is collected ethically, with proper consent, and that sensitive or personal information is handled appropriately.

The business goal identification phase is where the organization defines the purpose of the ML project and determines the compliance, regulatory, and legal requirements. These considerations must be addressed early in the lifecycle to ensure the solution adheres to applicable laws and standards. For example, in industries like finance or healthcare, this phase would identify data privacy regulations (e.g., GDPR, HIPAA) or fairness requirements that need to be incorporated into the ML workflow.