Amazon Bedrock Guardrails Quiz

What you'll be asked

1. What is the primary purpose of using Amazon Bedrock Guardrails?
2. You need to prevent your generative AI application from leaking sensitive customer data such as credit card numbers or social security numbers. Which Guardrail feature should you configure?
3. If you want to ensure that your model refuses to answer questions about specific topics like political opinions or competitor products, which configuration should you use?
4. A user attempts to manipulate your chatbot into ignoring its system instructions by using a 'jailbreak' attack. How does Amazon Bedrock Guardrails handle this?
5. You have integrated a Guardrail into your Bedrock application. If you have both a 'Content Filter' configured with a 'High' threshold and a 'Contextual Grounding' policy enabled, what happens when the model generates an answer that is factually unsupported by the provided source documents?

Answers and explanations come after you submit.