AIF-C01 Practice Test 25 — 15 Questions

Question 1

A company uses Amazon Comprehend to analyze customer sentiment. They discover the model performs poorly on text with slang from underrepresented groups. What is the most responsible action?

Accepted Answer

Collect more representative training data including slang. The core principle of responsible AI requires that models be trained on data that is representative of the populations they serve. Amazon Comprehend's sentiment analysis is a supervised machine learning model; its poor performance on slang from underrepresented groups indicates a training data bias. Collecting more representative training data, including that slang, directly addresses the root cause by enabling the model to learn the linguistic patterns of those groups, improving fairness and accuracy without restricting access or masking the problem.

Answer

Restrict model use to only standard English

Answer

Remove slang from input before inference

Answer

Adjust the confidence threshold only for those groups

Question 2

A healthcare company needs to use Amazon SageMaker Ground Truth for data labeling. The data includes protected health information (PHI) that must remain in the US. Which configuration meets the compliance requirements?

Accepted Answer

Use a private workforce consisting of the company's employees and launch the labeling job in the us-east-1 region. A private workforce consisting of the company's own employees ensures that PHI is never exposed to external workers, and launching the labeling job in us-east-1 keeps all data within the US, satisfying the data residency requirement. Amazon SageMaker Ground Truth allows you to restrict data access to a private workforce that you manage, and by selecting a US region, you ensure that data processing and storage remain within US borders.

Answer

Use a vendor-managed workforce and set up data encryption

Answer

Use a public workforce (Mechanical Turk) and select the US East region

Answer

Use a private workforce but launch the labeling job in the eu-west-1 region

Question 3

A company develops a chatbot using Amazon Lex. To ensure transparency, what should the chatbot do when it cannot answer a question?

Accepted Answer

Clearly state that it cannot answer and offer alternatives. Responsible AI guidelines, including those from AWS for Amazon Lex, require that when a chatbot cannot answer a question, it should clearly state its inability and offer alternatives (e.g., rephrasing the query or providing related topics). This maintains transparency and user trust, aligning with the 'Explainability' principle under the Guidelines for Responsible AI.

Answer

Remain silent and wait for the next input

Answer

Provide a random answer from a predefined list

Answer

Automatically escalate all unanswered questions to a human

Question 4

A data scientist trains a binary classification model and obtains the following results on the test set: accuracy 0.92, precision 0.90, recall 0.85, F1 0.87. The dataset has 5% positive class. The business requirement is to minimize false negatives. Which metric should the team prioritize?

Accepted Answer

Recall. Recall (sensitivity) measures the proportion of actual positives correctly identified, which directly addresses the business requirement to minimize false negatives. With a 5% positive class, accuracy is misleadingly high because the model can simply predict the majority class (negative) and still achieve 95% accuracy, but this would result in zero recall. Prioritizing recall ensures the model captures as many true positives as possible, reducing false negatives at the cost of potentially more false positives.

Answer

F1 score

Answer

Accuracy

Answer

Precision

Question 5

A financial institution is developing a model to detect fraudulent transactions. They want to ensure the model is robust and does not exhibit bias. Which TWO actions should they take?

Accepted Answer

Use Amazon SageMaker Clarify to compute fairness metrics like demographic parity. Option A is correct because Amazon SageMaker Clarify is the AWS service specifically designed to detect bias in data and models, and it computes fairness metrics such as demographic parity, disparate impact, and equal opportunity across sensitive groups, directly addressing the requirement to ensure the model does not exhibit bias. Option E is correct because a balanced training dataset that represents all transaction types and customer demographics reduces sampling bias and prevents the model from overfitting to majority groups, which is a foundational step for building a robust and fair fraud-detection model. Option B is incorrect because using a single unsegmented model for all customers can mask region-specific fraud patterns and amplify bias, rather than mitigate it. Option C is incorrect because raising the confidence threshold only tunes precision/recall trade-offs for false positives; it does nothing to detect or reduce bias. Option D is incorrect because encryption at rest and in transit is a security control for data protection and compliance, not a fairness or bias-mitigation measure.

Answer

Use a single model for all customers without segmenting by region

Answer

Deploy the model with a high confidence threshold to reduce false positives

Answer

Encrypt all transaction data at rest and in transit

Question 6

Refer to the exhibit. A developer is reviewing CloudWatch Logs for a deployed model and notices the same input appears multiple times with slightly different probabilities. What responsible AI concern does this pattern suggest?

Accepted Answer

The model is not robust; it produces inconsistent predictions for the same input.. The pattern of the same input producing slightly different probabilities indicates that the model's predictions are not deterministic for identical inputs. This violates the principle of robustness in responsible AI, which requires that a model should produce consistent outputs for the same input under the same conditions. Inconsistent predictions for identical inputs undermine trust and reliability, making the model non-robust.

Answer

The model is overfitting to the training data.

Answer

The model is exhibiting bias against a demographic group.

Answer

The input data is drifting from the training distribution.

Question 7

A company uses Amazon Bedrock Guardrails to filter harmful content. They want to ensure that the model does not generate responses containing specific keywords related to their internal project names. Which Guardrails component should they configure?

Accepted Answer

Word filters. Word filters in Amazon Bedrock Guardrails are specifically designed to block or mask responses containing custom keywords or phrases. By configuring word filters, the company can ensure that internal project names are not generated in model responses. This is the precise component for keyword-based filtering.

Answer

Harmful content filters

Answer

Topic restrictions

Answer

Grounding checks

Question 8

A team is using Amazon SageMaker to build a text classification model. They have raw text data in CSV files stored in Amazon S3. Before training, they need to perform feature engineering. Which THREE actions should they take? (Select THREE.)

Accepted Answer

Tokenize the text into words or subwords. For text classification, the raw text must be transformed into a machine-readable representation, so option B is correct because tokenizing the text into words or subwords is the essential first step that splits raw strings into discrete units a model can process. Option C is correct because removing punctuation and stop words reduces noise and dimensionality, helping the model focus on meaningful tokens rather than common, low-information words. Option D is correct because converting the tokenized text into numerical features via TF-IDF or word embeddings is required, since SageMaker training algorithms and models operate on numeric vectors, not raw strings. Option A is not appropriate here because normalizing numerical features to zero mean and unit variance applies to numeric columns, not raw text data. Option E is also not appropriate because one-hot encoding is used for categorical variables, which is not the primary transformation needed for free-form text in this scenario.

Answer

Normalize numerical features to have zero mean and unit variance

Answer

Encode categorical variables using one-hot encoding

Question 9

A research team is using Amazon SageMaker to fine-tune a large language model. They want to optimize training cost and time without sacrificing model quality. Which THREE strategies should they implement? (Choose 3)

Accepted Answer

Apply parameter-efficient fine-tuning (PEFT) techniques like LoRA.. Option B is correct because parameter-efficient fine-tuning techniques such as LoRA freeze most of the pretrained model weights and train only small low-rank adapter matrices, which drastically reduces GPU memory consumption, training time, and compute cost while preserving model quality. Option D is correct because SageMaker managed spot training can cut training costs by up to 90% versus on-demand instances, and pairing it with checkpointing to Amazon S3 lets training resume from the last checkpoint after a spot interruption rather than restarting from scratch. Option E is correct because mixed precision training with FP16 reduces memory footprint and leverages GPU tensor cores for faster matrix math, speeding up training and allowing larger effective batch sizes with minimal impact on model accuracy. Option A is not correct because simply moving to a larger multi-GPU instance raises cost and does not by itself optimize time or cost efficiency, and may be unnecessary if PEFT and FP16 already fit the workload. Option C is not correct because maximizing batch size to the GPU memory limit is not universally beneficial; overly large batches can hurt convergence and model quality and may require extensive hyperparameter retuning, so it is not a reliable cost-and-time optimization strategy.

Answer

Use a larger instance type with more GPUs.

Answer

Increase the batch size to the maximum that fits in GPU memory.

Question 10

Which THREE statements about Amazon SageMaker Ground Truth are correct? (Choose three.)

Accepted Answer

It provides built-in workflows for image classification and object detection.. Option B is correct because Amazon SageMaker Ground Truth ships with built-in labeling task templates and workflows for common job types, including image classification and object detection (as well as text, semantic segmentation, and bounding boxes). Option C is correct because Ground Truth supports automated data labeling, which uses active learning to have a model label high-confidence data while sending only low-confidence samples to human labelers, reducing cost and effort. Option D is correct because Ground Truth integrates directly with Amazon SageMaker, writing labeled datasets to Amazon S3 in augmented manifest format that SageMaker training jobs can consume. Option A is wrong because Ground Truth handles images, video, text, and 3D point clouds, not just text. Option E is wrong because Ground Truth supports multiple workforces, including private workforces, vendor-managed workforces, and Amazon Mechanical Turk, so it is not limited to a public Mechanical Turk workforce.

Answer

It can only be used for text data.

Answer

It can only use a public workforce from Amazon Mechanical Turk.

Question 11

A hospital is deploying an AI system to assist in diagnosing diseases from medical images. According to the EU AI Act, this system may be classified as high-risk. Which THREE requirements should the hospital address to comply with the EU AI Act for high-risk AI systems?

Accepted Answer

Ensure transparency and provision of information to users. The EU AI Act requires high-risk systems to have risk management, transparency, and human oversight among other requirements.

Answer

Use only open-source models

Answer

Deploy the system without any testing in a sandbox environment

Question 12

A research lab uses Amazon SageMaker to train a deep learning model for medical diagnosis. They need to ensure the model's decisions are interpretable to clinicians. Which SageMaker feature provides local and global feature importance?

Accepted Answer

SageMaker Clarify. SageMaker Clarify is the correct answer because it is specifically designed to provide both local and global feature importance for machine learning models. Local feature importance explains individual predictions (e.g., why a specific patient was diagnosed), while global feature importance shows which features most influence the model overall. This directly supports interpretability for clinicians, as required in the question.

Answer

SageMaker Model Monitor

Answer

SageMaker Experiments

Answer

SageMaker Debugger

Question 13

A developer is using Amazon Bedrock Agents to build a multi-step reasoning bot that can query a SQL database and summarize results. Which service should be integrated as the action group's Lambda function to execute SQL queries?

Accepted Answer

Amazon Aurora. Amazon Aurora is correct because it is a fully managed SQL-compatible relational database that can be queried using standard SQL. The Lambda function in the action group can use the boto3 library to execute SQL queries against Aurora, enabling the Bedrock Agent to retrieve and summarize data from a SQL database.

Answer

Amazon DynamoDB

Answer

Amazon OpenSearch Serverless

Answer

Amazon S3

Question 14

A healthcare company must train a model on sensitive patient data while complying with privacy regulations. They want to add noise to the training process to prevent re-identification. Which technique should they implement?

Accepted Answer

Differential privacy. Differential privacy is the correct technique because it adds calibrated noise to the training process (e.g., via gradient clipping and noise injection in stochastic gradient descent) to ensure that the model's outputs do not reveal whether any individual's data was included in the training set. This provides a formal mathematical guarantee (ε-differential privacy) that limits the risk of re-identification, which is essential for complying with privacy regulations like HIPAA or GDPR when training on sensitive patient data.

Answer

k-anonymity

Answer

Federated learning

Answer

Homomorphic encryption

Question 15

An e-commerce company uses Amazon Bedrock to generate product descriptions. They notice the descriptions are too long and contain repetitive phrases. Which parameter adjustment can help?

Accepted Answer

Increase frequency penalty. Increasing the frequency penalty reduces the likelihood of the model repeating the same phrases or tokens, directly addressing the issue of repetitive language in generated product descriptions. This parameter penalizes tokens that have already appeared in the text, encouraging more diverse output and naturally shortening overly long descriptions by avoiding redundant loops.

Answer

Increase temperature

Answer

Increase top_p

Answer

Decrease presence penalty