Guidelines for Responsible AI
Explainability and Transparency
Explainability helps people understand model predictions or outputs, while transparency communicates intended use, limitations, and evaluation results. AWS AI Service Cards and model documentation are important responsible AI artifacts.
Concepts
- SageMaker Clarify helps detect bias and explain feature importance.
- AWS AI Service Cards document intended use cases, limitations, responsible AI considerations, and performance factors.
- Model cards, data documentation, and human-readable explanations support responsible adoption.
Exam tips
- SageMaker Clarify provides bias and feature-importance explainability capabilities.
- AWS AI Service Cards document intended use, limitations, and responsible AI considerations.
- Model cards and data documentation support informed review and governance.