Free AWS-CERTIFIED-AI-PRACTITIONER-AIF-C01 – Amazon AWS-CERTIFIED-AI-PRACTITIONER-AIF-C01 Practice Test Question

Reference / correct answer:

Most accepted answer: D. Jailbreak

Community votes: D=6

Selected Answer: D D is the correct answer upvoted 1 times

Selected Answer: D Explicação: Jailbreak é uma técnica usada para burlar ou contornar as restrições de segurança impostas a um modelo de linguagem (como filtros de conteúdo ou limites de comportamento). O objetivo é fazer com que o modelo gere respostas que normalmente seriam bloqueadas, como conteúdo prejudicial, ofensivo ou perigoso. Esse tipo de teste é comum em avaliações de segurança de LLMs, especialmente em ambientes controlados de validação e pesquisa. upvoted 1 times

Selected Answer: D Jailbreaking refers to bypassing or disabling the security restrictions placed on a system—in this case, a foundation model (FM)—to make the system behave in unintended ways, often to produce harmful or malicious content. In the context of AI, jailbreaking typically involves manipulating the model's behavior or output by exploiting vulnerabilities in its design or safety features. upvoted 4 times

Selected Answer: D The correct answer is D. A jailbreak is an attempt to bypass an AI model's built-in safety controls. upvoted 1 times

Selected Answer: D D. Jailbreak Explanation: Jailbreaking is a technique used to bypass the safety features and restrictions of a foundation model (FM). The goal is to manipulate the model into generating harmful, inappropriate, or otherwise unintended content, despite the safeguards in place. This is often done to test the robustness of the model's safety mechanisms. upvoted 2 times

Reference / correct answer:

Most accepted answer: A. Amazon SageMaker Clarify

Community votes: A=6

Selected Answer: A A. Amazon SageMaker Clarify upvoted 1 times

Selected Answer: A A is the correct answer upvoted 1 times

Selected Answer: A Sagemaker Clarify. But here's how you think about it. Think of a dataset as a big Excel sheet. It's made up of columns and rows. A model uses the all but one of the columns to explain the value in the last column. So if you want clarity on how it works, you might need to understand how the columns predict the last column (model explainability / feature attribution), and also if the rows have enough data for each subgroup (fairness / bias). tl;dr - Sagemaker Clarify will clarify how the columns and rows fit the model. upvoted 1 times

Selected Answer: A The solution that meets the requirements of detecting bias and explaining model predictions in a loan approval scenario is A. Amazon SageMaker Clarify upvoted 3 times

Selected Answer: A The correct answer is A. SageMaker Clarify provides both bias detection and model explainability features. upvoted 2 times

Reference / correct answer:

Most accepted answer: A. AWS Key Management Service (AWS KMS)

Community votes: A=5

Selected Answer: A The company needs to use a company-managed encryption key to encrypt model artifacts. This points directly to key management. A. AWS Key Management Service (AWS KMS): This is the correct answer. AWS KMS allows you to create and manage encryption keys, including customer-managed keys (CMKs), which give you control over the key lifecycle and usage. B. Amazon Inspector: Inspector is a vulnerability management service that scans for security vulnerabilities in your AWS resources. C. Amazon Macie: Macie is a data security and privacy service that uses machine learning to discover and protect sensitive data in AWS. D. AWS Secrets Manager: Secrets Manager helps you manage secrets such as passwords, API keys, and database credentials. While it can store encrypted secrets, it's not the primary service for managing encryption keys used to protect model artifacts at rest. upvoted 5 times

Selected Answer: A A. AWS Key Management Service (AWS KMS) upvoted 1 times

Selected Answer: A A is the correct answer upvoted 1 times

Selected Answer: A Amazon Bedrock supports encryption of model artifacts using AWS Key Management Service (AWS KMS). AWS KMS allows you to use a company-managed encryption key (customer-managed key or CMK) to encrypt the model artifacts created during model customization jobs. upvoted 2 times

The company needs to use a company-managed encryption key to encrypt model artifacts. This points directly to key management. A. AWS Key Management Service (AWS KMS): This is the correct answer. AWS KMS allows you to create and manage encryption keys, including customer-managed keys (CMKs), which give you control over the key lifecycle and usage. B. Amazon Inspector: Inspector is a vulnerability management service that scans for security vulnerabilities in your AWS resources. C. Amazon Macie: Macie is a data security and privacy service that uses machine learning to discover and protect sensitive data in AWS. D. AWS Secrets Manager: Secrets Manager helps you manage secrets such as passwords, API keys, and database credentials. While it can store encrypted secrets, it's not the primary service for managing encryption keys used to protect model artifacts at rest. upvoted 1 times