Responsible AI Tools & Practices Match
MediumRace the clock — pair each term with its definition as fast as you can, then try to beat your best time. A quick, replayable warm-up for this set.
Shuffling the deck…
Terms in this set
- Explainability The ability to explain why a model produced a particular output or decision.
- Guardrails Guardrails for Amazon Bedrock — configurable safety filters for FM inputs and outputs.
- Clarify Amazon SageMaker Clarify — detects bias in data and models and explains predictions.
- Model Monitor Amazon SageMaker Model Monitor — detects data and model quality drift in production.
- Model Card A document recording a model's intended use, performance, and limitations.
- Toxicity Harmful, offensive, or abusive content in model output.
- Human-in-the-Loop Inserting human review into an AI workflow for oversight on important decisions.