Courseiva

AWS Certified AI Practitioner AIF-C01 (AIF-C01) — Questions 451–500

500 questions total · 7pages · All types, answers revealed

Take a mock examExam hub
Page 6

Page 7 of 7

451
MCQhard

A team is using Amazon Bedrock to generate images from text prompts. The generated images often contain artifacts and do not match the prompt description. Which combination of steps should the team take to improve image quality?

A.Fine-tune the model using SageMaker Ground Truth and increase the training epochs.
B.Increase the max token count and use a larger model variant.
C.Refine the prompt with more descriptive language and adjust the CFG scale and inference steps.
D.Use a different foundation model and increase the image resolution.
AnswerC

Better prompts and tuning inference parameters directly improve image quality.

Why this answer

Option C is correct because refining the prompt with more descriptive language helps the model better interpret the user's intent, while adjusting the CFG (Classifier-Free Guidance) scale controls how strictly the model adheres to the prompt, and increasing inference steps allows the diffusion process to produce higher-quality, artifact-free images. These are standard hyperparameters in diffusion-based image generation models on Amazon Bedrock, directly addressing both artifacts and prompt mismatch.

Exam trap

AWS often tests the misconception that image quality issues are best solved by model retraining or changing the model, rather than by adjusting inference-time parameters like CFG scale and inference steps, which are the immediate and correct levers for prompt adherence and artifact reduction.

How to eliminate wrong answers

Option A is wrong because fine-tuning a model using SageMaker Ground Truth and increasing training epochs is a data labeling and retraining approach that is overkill and not directly applicable to improving inference-time image quality for a pre-trained Bedrock model; it also does not address prompt adherence or artifact reduction. Option B is wrong because increasing the max token count and using a larger model variant does not fix artifacts or prompt mismatch—max token count affects text generation length, not image quality, and a larger model may not inherently improve prompt alignment without prompt engineering. Option D is wrong because using a different foundation model and increasing image resolution may change output characteristics but does not systematically address artifacts or prompt mismatch; higher resolution can even amplify artifacts if the underlying generation process is not optimized.

Full explanation →
452
MCQeasy

A data scientist is using Amazon SageMaker to train a large language model from scratch. Which AWS service is most suitable for managing the training infrastructure, including automatic scaling and spot instance recovery?

A.AWS Lambda function.
B.Amazon SageMaker Notebook instance.
C.Amazon SageMaker Training job.
D.Amazon EC2 with a custom setup.
AnswerC

SageMaker Training manages infrastructure, automatically recovers from spot interruptions, and scales.

Why this answer

Amazon SageMaker Training jobs are the most suitable service for managing training infrastructure because they provide built-in automatic scaling, managed spot instance recovery, and distributed training orchestration. This allows the data scientist to focus on model development rather than provisioning and managing EC2 instances, load balancers, or recovery scripts.

Exam trap

Cisco often tests the distinction between managed services (SageMaker Training) and unmanaged services (EC2 custom setup), where candidates mistakenly choose EC2 thinking they need full control, overlooking SageMaker's built-in spot recovery and scaling capabilities.

How to eliminate wrong answers

Option A is wrong because AWS Lambda functions are serverless compute services designed for short-running, event-driven tasks (max 15-minute execution time) and cannot manage long-running training jobs or infrastructure scaling. Option B is wrong because Amazon SageMaker Notebook instances are interactive development environments for prototyping and exploration, not designed to manage production training infrastructure or handle automatic scaling and spot instance recovery. Option D is wrong because Amazon EC2 with a custom setup requires manual provisioning, configuration of auto-scaling groups, and custom scripts for spot instance interruption handling, which is less efficient and more error-prone than SageMaker's managed training service.

Full explanation →
453
MCQhard

Refer to the exhibit. A SageMaker real-time endpoint is experiencing increasing latency and memory errors after running for a few hours. What is the most likely cause and recommended fix?

A.Scale the endpoint to a larger instance type, such as ml.r5.large
B.Enable auto-scaling to add instances during high load
C.Use SageMaker Debugger to identify and fix a memory leak in the inference code
D.Use SageMaker Model Monitor to detect data drift
AnswerC

The increasing memory usage over time indicates a leak; Debugger can help identify the issue.

Why this answer

Option C is correct because the symptoms—increasing latency and memory errors after running for a few hours—point to a memory leak in the inference code. SageMaker Debugger can monitor system metrics like memory utilization and detect anomalies, helping to identify the root cause of the leak. Fixing the memory leak directly resolves the progressive degradation, whereas scaling or auto-scaling only masks the symptom.

Exam trap

Cisco often tests the distinction between scaling solutions (which address capacity) and debugging tools (which address code defects), trapping candidates who confuse symptom relief with root cause resolution.

How to eliminate wrong answers

Option A is wrong because scaling to a larger instance type (e.g., ml.r5.large) provides more memory but does not address the underlying memory leak; the leak will eventually exhaust the larger memory pool as well. Option B is wrong because enabling auto-scaling adds more instances to handle load, but it does not fix the memory leak in the inference code; each instance will still experience the same progressive memory exhaustion. Option D is wrong because SageMaker Model Monitor detects data drift (changes in input data distribution), not memory leaks or latency issues caused by code defects.

Full explanation →
454
MCQeasy

A marketing agency uses a foundation model to generate images for social media campaigns. Some generated images have contained violent or inappropriate content, damaging the brand. The agency needs to prevent such content from being displayed automatically. They are using Amazon Bedrock for image generation with Stable Diffusion. What is the most effective way to filter out inappropriate images?

A.Use Amazon Rekognition to analyze images after generation.
B.Manually review all images before posting.
C.Restrict the prompt to avoid triggering keywords.
D.Enable the safety checker in Amazon Bedrock's image generation models.
AnswerD

Built-in safety checker filters out inappropriate images without additional overhead.

Why this answer

Option B is correct because Stable Diffusion models in Bedrock include a safety checker that can detect and block NSFW content before output. Option A (Amazon Rekognition) introduces additional cost and latency. Option C (manual review) is not scalable.

Option D (restrict prompt) is unreliable as the model can still generate inappropriate content from safe prompts.

Full explanation →
455
MCQhard

An e-commerce company is using a foundation model to generate product descriptions. They want to reduce costs by caching frequently requested descriptions. Which AWS service should they use to implement a cache?

A.Amazon CloudFront
B.Amazon DynamoDB
C.Amazon S3
D.Amazon ElastiCache
AnswerD

ElastiCache provides low-latency caching for frequently used data.

Why this answer

Amazon ElastiCache is the correct choice because it provides an in-memory caching layer (using Redis or Memcached) that can store frequently requested product descriptions, reducing the need to invoke the foundation model repeatedly. This directly lowers inference costs and latency by serving cached responses instead of generating new ones each time.

Exam trap

Cisco often tests the distinction between caching at the application layer (ElastiCache) versus caching at the content delivery layer (CloudFront), leading candidates to mistakenly choose CloudFront for any caching need.

How to eliminate wrong answers

Option A is wrong because Amazon CloudFront is a content delivery network (CDN) that caches static and dynamic content at edge locations, but it is not designed for application-level caching of model-generated text; it caches HTTP responses, not arbitrary key-value data. Option B is wrong because Amazon DynamoDB is a fully managed NoSQL database optimized for high-throughput, low-latency reads and writes, but it is not a caching service; using it as a cache would incur higher costs and lack native TTL-based eviction policies for transient data. Option C is wrong because Amazon S3 is an object storage service for storing large amounts of unstructured data, not a low-latency cache; retrieving descriptions from S3 would introduce significant latency compared to an in-memory cache, defeating the purpose of cost reduction.

Full explanation →
456
MCQmedium

A company uses Amazon Bedrock to generate marketing content. They want to reduce costs while maintaining response quality. Which action is most effective?

A.Fine-tune a larger model to improve accuracy and reduce retries.
B.Increase the temperature parameter to get shorter responses.
C.Select a smaller foundation model that still meets accuracy requirements.
D.Cache previous responses to reuse for similar prompts.
AnswerC

Smaller models have lower per-token costs and are faster.

Why this answer

Option D is correct because selecting a smaller, efficient foundation model can reduce cost per token while maintaining quality for simple tasks. Option A is wrong because increasing temperature does not reduce cost. Option B is wrong because caching may not be effective for variable outputs.

Option C is wrong because fine-tuning increases cost.

Full explanation →
457
MCQhard

A financial institution is deploying a fraud detection model using Amazon SageMaker. The model must be able to handle sudden spikes in inference requests during promotional events while keeping costs low. The team wants to use a serverless architecture to avoid provisioning idle capacity and to scale automatically from zero. However, the inference latency requirement is under 5 seconds for each request. Which SageMaker inference option should they choose?

A.Use Amazon SageMaker Serverless Inference
B.Use Amazon SageMaker Multi-Model Endpoints
C.Use Amazon SageMaker real-time endpoints with auto-scaling
D.Use Amazon SageMaker Asynchronous Inference
AnswerA

Serverless Inference scales automatically from zero and reduces costs during idle periods.

Why this answer

Amazon SageMaker Serverless Inference is the correct choice because it automatically scales from zero to handle sudden spikes in inference requests, aligning with the requirement to avoid provisioning idle capacity. It also meets the sub-5-second latency requirement for fraud detection, as it is designed for low-latency, on-demand inference without managing underlying infrastructure.

Exam trap

AWS often tests the misconception that serverless inference cannot meet low-latency requirements, but SageMaker Serverless Inference is specifically designed for sub-second to few-second latency, making it suitable for real-time fraud detection scenarios.

How to eliminate wrong answers

Option B is wrong because Multi-Model Endpoints require provisioned instances and do not scale from zero; they are designed to host multiple models on a single endpoint but still incur costs for idle capacity. Option C is wrong because real-time endpoints with auto-scaling still require a baseline of provisioned instances, which can lead to idle capacity costs during low-traffic periods, and they do not scale from zero. Option D is wrong because Asynchronous Inference is intended for large payloads and longer processing times (typically minutes), not for sub-5-second latency requirements, and it queues requests rather than providing real-time responses.

Full explanation →
458
MCQhard

A healthcare organization wants to use generative AI to draft clinical notes from patient-physician conversations. They must comply with HIPAA and minimize false medical information. Which approach should they take?

A.Use Amazon SageMaker JumpStart with a publicly available clinical model and no additional modifications.
B.Use a generic open-source LLM hosted on Amazon EC2 with manual prompt engineering.
C.Use Amazon Bedrock with a HIPAA-eligible foundation model and connect it to a medical knowledge base via RAG.
D.Use Amazon Bedrock with a large foundation model and a high temperature setting for creativity.
AnswerC

Ensures compliance and accuracy through grounding on trusted medical sources.

Why this answer

Option A is correct because Amazon Bedrock offers HIPAA-eligible models and allows grounding with medical knowledge bases to reduce hallucinations. Option B is wrong because it does not use grounding. Option C is wrong because open-source LLMs may not be HIPAA-compliant.

Option D is wrong because increasing temperature introduces more randomness, worsening accuracy.

Full explanation →
459
Multi-Selectmedium

Which THREE of the following are factors to consider when selecting a foundation model for a text generation task?

Select 3 answers
A.Supported output modalities
B.Pricing per token
C.Model size (parameters)
D.Training data source and diversity
E.Availability of automatic scaling
AnswersB, C, D

Cost per token affects operational expense.

Why this answer

Pricing per token is a critical factor because foundation model APIs (e.g., Amazon Bedrock, OpenAI) charge based on the number of input and output tokens. For text generation tasks, token costs directly impact operational budgets, especially for high-volume or long-context applications. Selecting a model with lower per-token pricing can significantly reduce inference costs without sacrificing quality.

Exam trap

AWS often tests the distinction between model-level attributes (e.g., token pricing, training data, parameter count) and platform-level operational features (e.g., scaling, output modalities), leading candidates to incorrectly select options like automatic scaling or multimodal support for a text-only task.

Full explanation →
460
MCQmedium

A company uses Amazon Bedrock to generate summarizations of lengthy reports. Users report that the summaries are too verbose and include excessive detail. Which prompt engineering technique should the team apply to address this issue?

A.Reduce the input context length to limit available information.
B.Increase the maxTokens parameter in the inference request.
C.Include few-shot examples of desired outputs.
D.Add explicit constraints like 'Provide a concise summary in two sentences.'
AnswerD

Explicit constraints directly guide the model to produce shorter output, addressing verbosity effectively.

Why this answer

Option D is correct because adding explicit constraints like 'Provide a concise summary in two sentences' directly instructs the model to limit verbosity and detail. This prompt engineering technique uses clear, specific instructions to control output length and style, which is the most effective way to address overly verbose summaries without altering model parameters or input data.

Exam trap

The trap here is that candidates confuse reducing input length (Option A) with controlling output length, or they mistakenly think increasing maxTokens (Option B) can somehow shorten output, when in fact it does the opposite.

How to eliminate wrong answers

Option A is wrong because reducing input context length does not guarantee concise output; the model may still generate verbose summaries from the remaining text, and it risks losing critical information needed for accurate summarization. Option B is wrong because increasing the maxTokens parameter actually allows the model to generate longer outputs, which would exacerbate the verbosity issue rather than solve it. Option C is wrong because few-shot examples can guide output format but are less direct and reliable than explicit constraints; they may not consistently enforce conciseness, especially if the examples themselves are not perfectly aligned with the desired brevity.

Full explanation →
461
MCQeasy

A team is evaluating a classification model. The confusion matrix shows: TP=80, FN=20, FP=10, TN=90. What is the precision?

A.0.89
B.0.75
C.0.80
D.0.90
AnswerA

Precision = 80/(80+10) = 0.8889 ≈ 0.89.

Why this answer

Precision is calculated as TP / (TP + FP). Here, TP=80 and FP=10, so precision = 80 / (80 + 10) = 80 / 90 = 0.888..., which rounds to 0.89. This metric measures the proportion of positive identifications that were actually correct.

Exam trap

Cisco often tests the distinction between precision and recall by providing confusion matrix values that make one metric easy to miscalculate if you confuse the denominator (TP+FP vs TP+FN).

How to eliminate wrong answers

Option B (0.75) is wrong because it incorrectly uses FN in the denominator, likely confusing precision with recall (TP / (TP + FN)). Option C (0.80) is wrong because it uses only TP divided by the total number of actual positives (TP + FN), which is recall, not precision. Option D (0.90) is wrong because it uses TN in the denominator or calculates accuracy (TP + TN) / total, which is not precision.

Full explanation →
462
MCQmedium

A large enterprise uses Amazon Bedrock to power a conversational agent that handles customer service inquiries. The agent is built using Bedrock Agents and retrieves information from a knowledge base that contains product documentation and FAQs. Recently, users have reported that the agent sometimes provides incorrect information that contradicts the knowledge base. The development team verified that the knowledge base contains accurate and up-to-date data. They also confirmed that the retrieval process correctly fetches relevant documents. However, the agent occasionally ignores the retrieved context and generates plausible-sounding but incorrect answers. The team is concerned about customer trust and wants to improve the accuracy of the agent's responses without overhauling the architecture. They have already tuned the prompt template to instruct the model to use the context. The issue persists. Which additional action should the team take to reduce the number of hallucinated responses?

A.Reduce the chunk size of documents in the knowledge base to retrieve more granular information.
B.Switch to a larger foundation model with more parameters.
C.Increase the temperature parameter of the foundation model.
D.Add explicit instructions in the system prompt to require the model to base its answers solely on the retrieved context and to state when it doesn't have enough information.
AnswerD

Correct: Strengthening the prompt with explicit directives can reduce hallucinations by forcing the model to rely on the provided context.

Why this answer

Option B directly addresses the model ignoring context by strengthening the instruction. Option A increases randomness, Option C does not guarantee use of context, Option D may not help if retrieval is already good.

Full explanation →
463
MCQeasy

A startup is developing a mobile app that uses facial recognition to verify user identity for account access. The app is intended for a global audience, but the training data predominantly includes images of light-skinned individuals. During beta testing, users with darker skin tones report frequent verification failures, while light-skinned users have a high success rate. The startup wants to release the app soon and needs to address this fairness issue without delaying the launch too much. The team has limited resources. Which approach should they take to most effectively mitigate the bias while meeting the launch timeline?

A.Apply a post-processing rule to increase acceptance rate for users with darker skin tones
B.Lower the similarity threshold for all users to improve acceptance rates
C.Defer verification for users with darker skin tones to manual human review
D.Collect more diverse training data and augment the existing dataset, then retrain the model
AnswerD

Adding diverse data addresses the root cause of bias.

Why this answer

The most effective approach is to collect additional training data representing diverse skin tones and augment the dataset, then retrain the model. This directly addresses the data imbalance. Applying a post-processing rule without retraining may not fix the underlying model bias.

Deferring to humans is a temporary workaround and does not scale. Reducing the threshold for all users could increase false positives and may not be acceptable.

Full explanation →
464
MCQeasy

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

A.Remain silent and wait for the next input
B.Provide a random answer from a predefined list
C.Clearly state that it cannot answer and offer alternatives
D.Automatically escalate all unanswered questions to a human
AnswerC

Honest communication builds trust.

Why this answer

Option C is correct because 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.

Exam trap

AWS often tests the misconception that a chatbot should always escalate or remain passive when it cannot answer, but the correct approach under responsible AI is to acknowledge the limitation and offer alternatives, not to hide or mislead.

How to eliminate wrong answers

Option A is wrong because remaining silent and waiting for the next input violates transparency and can confuse users, as it provides no feedback or guidance. Option B is wrong because providing a random answer from a predefined list is deceptive and can mislead users, undermining the principle of honesty and accountability in AI. Option D is wrong because automatically escalating all unanswered questions to a human is inefficient and not always necessary; the chatbot should first attempt to offer alternatives or clarify before escalation, as per responsible AI practices.

Full explanation →
465
Multi-Selecteasy

A company uses Amazon Bedrock to build a generative AI application. They need to secure the application by restricting access to the model and preventing sensitive data from being stored in prompts. Which TWO actions should they take? (Choose two.)

Select 2 answers
A.Enable VPC Flow Logs for Bedrock.
B.Enable AWS CloudTrail to log all Bedrock API calls.
C.Use Amazon Bedrock Guardrails to define content filters and deny topics.
D.Implement IAM policies that allow only specific users to invoke the model.
E.Use AWS KMS to encrypt the model data.
AnswersC, D

Guardrails can filter sensitive data and enforce content policies.

Why this answer

Amazon Bedrock Guardrails (Option C) directly addresses the need to prevent sensitive data from being stored in prompts by allowing you to define content filters, deny topics, and configure sensitive information filters that block or mask such data before it reaches the model. This is the native service feature designed for content safety and data leakage prevention.

Exam trap

AWS often tests the distinction between logging/auditing services (CloudTrail, VPC Flow Logs) and actual security enforcement mechanisms (Guardrails, IAM), leading candidates to confuse observability with prevention.

Full explanation →
466
MCQmedium

Refer to the exhibit. A security officer has attached this IAM policy to a user to allow invocation of a SageMaker endpoint. However, the user is unable to invoke the endpoint from an EC2 instance in the same VPC. What is the most likely cause?

A.The SageMaker endpoint is not configured to accept traffic from the VPC.
B.The IAM policy does not allow the sagemaker:InvokeEndpoint action without conditions.
C.The user's IAM role is missing the s3:GetObject permission for model artifacts.
D.The EC2 instance is not using the specified VPC endpoint (vpce-12345678).
AnswerD

The condition restricts invocation to traffic coming through that specific VPC endpoint.

Why this answer

The policy condition requires that the request originates from a specific VPC endpoint (vpce-12345678). If the EC2 instance is not using that endpoint (e.g., it is using a different endpoint or direct VPC access), the request will be denied. Option B is correct because the instance likely does not use that specific VPC endpoint.

Full explanation →
467
Multi-Selecteasy

Which TWO AWS services can be used together to build a chatbot that leverages a foundation model for natural language understanding?

Select 2 answers
A.Amazon Rekognition
B.Amazon Lex
C.Amazon Polly
D.AWS Glue
E.Amazon Bedrock
AnswersB, E

Lex handles dialog management and intent recognition.

Why this answer

Amazon Lex provides the conversational interface and natural language understanding (NLU) to interpret user intents and manage dialog, while Amazon Bedrock gives access to foundation models (FMs) for advanced natural language generation and understanding. Together, Lex can route utterances to a Bedrock FM via a Lambda function or direct integration, enabling a chatbot that leverages a pre-trained FM for richer responses.

Exam trap

AWS often tests the distinction between services that handle conversational interfaces (Lex) versus those that provide generative AI models (Bedrock), tempting candidates to pick Polly (speech output) or Rekognition (vision) as part of a chatbot, when they are not core to NLU or FM integration.

Full explanation →
468
MCQhard

A team is training a deep learning model using Horovod distributed training on SageMaker. They observe that the loss stops decreasing after a few epochs. Which technique should they implement to reduce overfitting?

A.Increase learning rate
B.Add more layers to the model
C.Reduce the number of epochs
D.Use dropout regularization
AnswerD

Dropout is a regularization technique that reduces overfitting.

Why this answer

Dropout regularization randomly drops a fraction of neurons during training, which prevents the model from relying too heavily on specific features and forces it to learn more robust representations. This directly addresses overfitting, which is the likely cause of the loss plateauing after a few epochs in a Horovod distributed training setup on SageMaker.

Exam trap

Cisco often tests the misconception that early stopping (reducing epochs) is a regularization technique to reduce overfitting, but the trap here is that early stopping only halts training and does not actively prevent the model from memorizing noise during the epochs it does train.

How to eliminate wrong answers

Option A is wrong because increasing the learning rate can cause the optimizer to overshoot minima, leading to divergence or unstable training, not a reduction in overfitting. Option B is wrong because adding more layers increases model capacity, which typically worsens overfitting by making it easier for the model to memorize noise. Option C is wrong because reducing the number of epochs only stops training earlier; it does not address the underlying overfitting issue, and the loss may still plateau due to memorization rather than convergence.

Full explanation →
469
MCQeasy

A company uses Amazon Bedrock Agents to build an agent that interacts with users through a chat interface. The agent is configured with a knowledge base containing product documentation. Sometimes the agent fails to answer simple questions like 'What is your return policy?' and instead says it cannot find the answer. The knowledge base does contain the return policy. What is the most likely reason?

A.Increase the agent's maximum timeout for processing
B.Use a more powerful foundation model for reasoning
C.Add more documents to the knowledge base
D.Simplify and clarify the agent's instruction prompt to emphasize knowledge base usage
AnswerD

A clear prompt instructing the agent to consult the knowledge base for all answers can dramatically improve consistency.

Why this answer

The agent's instruction prompt might be too complex or not explicitly directing the agent to use the knowledge base. Simplifying the prompt to clearly instruct the agent to first search the knowledge base can resolve the issue. Increasing timeout or adding more data is unnecessary.

A stronger model may help but is not the root cause.

Full explanation →
470
MCQhard

A team is fine-tuning a foundation model using SageMaker. They want to minimize training time while keeping the model's original knowledge. Which technique is BEST suited?

A.Use Parameter Efficient Fine-Tuning (PEFT) such as LoRA
B.Use distributed training across multiple GPUs
C.Use prompt engineering instead of fine-tuning
D.Full fine-tuning on the new dataset
AnswerA

PEFT methods adapt the model with fewer trainable parameters, reducing training time and preserving original knowledge.

Why this answer

Parameter Efficient Fine-Tuning (PEFT) methods like LoRA (Low-Rank Adaptation) are best suited because they freeze the pre-trained model weights and inject trainable low-rank matrices into specific layers, drastically reducing the number of trainable parameters. This minimizes training time and computational cost while preserving the model's original knowledge, as only a small fraction of parameters are updated during fine-tuning.

Exam trap

AWS often tests the distinction between techniques that modify the model (fine-tuning) versus those that only change the input (prompt engineering), and the trap here is that candidates may choose distributed training (Option B) thinking it reduces time, but it does not address parameter efficiency or knowledge preservation as directly as PEFT.

How to eliminate wrong answers

Option B is wrong because distributed training across multiple GPUs accelerates training but does not inherently preserve the model's original knowledge or reduce the number of updated parameters; it still requires full or partial parameter updates and does not address the goal of minimizing training time through parameter efficiency. Option C is wrong because prompt engineering is a zero-shot or few-shot inference technique that does not involve training at all, so it cannot be used to fine-tune the model on a new dataset. Option D is wrong because full fine-tuning updates all model parameters, which is computationally expensive, time-consuming, and risks catastrophic forgetting of the original knowledge, contrary to the goal of minimizing training time while preserving original knowledge.

Full explanation →
471
MCQeasy

A startup with limited ML expertise wants to quickly prototype a binary classification model using a small customer dataset. They need a managed environment to run Jupyter notebooks and access pre-built algorithms. Which AWS service should they choose?

A.AWS Lambda
B.Amazon SageMaker
C.Amazon EMR
D.AWS Glue
AnswerB

SageMaker provides managed notebooks and built-in algorithms for quick experimentation.

Why this answer

Amazon SageMaker is the correct choice because it provides a fully managed environment for Jupyter notebooks and includes built-in, pre-built algorithms for binary classification. This allows the startup to quickly prototype without deep ML expertise, as SageMaker handles infrastructure, scaling, and model training.

Exam trap

Cisco often tests the distinction between managed ML platforms (SageMaker) and general-purpose compute or data processing services (Lambda, EMR, Glue), leading candidates to pick a service that can run code but lacks the specific notebook and pre-built algorithm capabilities required.

How to eliminate wrong answers

Option A is wrong because AWS Lambda is a serverless compute service for running code in response to events, not a managed environment for Jupyter notebooks or pre-built ML algorithms. Option C is wrong because Amazon EMR is a big data processing service using frameworks like Apache Spark and Hadoop, not designed for interactive Jupyter notebook-based ML prototyping with pre-built algorithms. Option D is wrong because AWS Glue is a serverless data integration and ETL service, not a platform for running Jupyter notebooks or accessing pre-built ML models.

Full explanation →
472
MCQmedium

A media company runs batch inference jobs to generate captions for thousands of images weekly using a foundation model on Amazon Bedrock. They want to minimize costs while maintaining predictable throughput. Which pricing option should they choose?

A.SageMaker Batch Transform
B.On-demand inference
C.Provisioned Throughput
D.Spot instances (EC2 Spot)
AnswerC

Reserves capacity for a model, providing consistent performance and lower per-token cost for large batches.

Why this answer

Provisioned Throughput reserves capacity for a specific model, offering predictable performance and cost savings for steady workloads. On-demand is pay-per-use but may be costlier for high volume. Batch Transform is for SageMaker, not Bedrock.

Spot instances are not available for Bedrock.

Full explanation →
473
MCQmedium

A data scientist is building a binary classification model for fraud detection. The dataset is highly imbalanced (99% legitimate, 1% fraud). Which metric is most appropriate to evaluate model performance?

A.Accuracy
B.F1-score
C.Recall
D.Precision
AnswerB

F1-score is the harmonic mean of precision and recall, providing a balanced evaluation for imbalanced datasets.

Why this answer

In highly imbalanced datasets (99% legitimate, 1% fraud), accuracy is misleading because a model that predicts all transactions as legitimate would achieve 99% accuracy but fail to detect any fraud. The F1-score is the harmonic mean of precision and recall, providing a balanced measure that accounts for both false positives and false negatives, making it the most appropriate metric for evaluating fraud detection models.

Exam trap

Cisco often tests the misconception that accuracy is always the best metric, especially when candidates overlook the impact of class imbalance on model evaluation.

How to eliminate wrong answers

Option A is wrong because accuracy is not suitable for imbalanced datasets; it can be artificially high (e.g., 99%) even if the model never identifies any fraud, as it only measures overall correct predictions. Option C is wrong because recall alone (true positive rate) ignores false positives; a model that predicts all transactions as fraud would have perfect recall but terrible precision, leading to many false alarms. Option D is wrong because precision alone ignores false negatives; a model that predicts only one transaction as fraud (and gets it right) could have 100% precision but miss the other 99 fraud cases, failing the detection goal.

Full explanation →
474
Multi-Selecthard

A company is deploying an Amazon Lex chatbot that processes customer credit card information. The company must ensure that the chatbot does not log sensitive data and that the data is encrypted in transit. Which THREE actions should the company take? (Choose THREE.)

Select 3 answers
A.Ensure the Amazon Lex endpoint uses HTTPS for all API calls
B.Enable audio logging in Amazon Lex with content redaction enabled
C.Enable CloudTrail data events for the Lex bot to capture conversation logs
D.Enable S3 server access logs on the bucket storing the Lex bot configuration
E.Configure the Amazon Lex bot to use obfuscation for the credit card slot type
AnswersA, B, E

HTTPS encrypts data in transit.

Why this answer

Option A is correct because Amazon Lex endpoints support HTTPS, which encrypts data in transit using TLS. By ensuring all API calls use HTTPS, the company protects sensitive credit card information from being intercepted during communication between the client and the Lex service.

Exam trap

The trap here is that candidates may confuse CloudTrail data events (which log API activity) with conversation logging (which captures user utterances), leading them to incorrectly select option C as a solution for preventing sensitive data logging.

Full explanation →
475
MCQhard

A data scientist is fine-tuning a foundation model on a custom dataset using Amazon SageMaker. After training, the model shows high accuracy on training data but poor on validation. Which action should be taken?

A.Add dropout layers
B.Reduce training epochs or add regularization
C.Increase learning rate
D.Use a different foundation model
AnswerB

Reducing epochs prevents overfitting; regularization also helps.

Why this answer

The model is overfitting, as indicated by high training accuracy but poor validation performance. Reducing training epochs or adding regularization (e.g., L1/L2 weight decay) directly addresses overfitting by limiting the model's capacity to memorize noise. In Amazon SageMaker, this can be implemented via hyperparameter tuning or by modifying the training script to include regularization terms.

Exam trap

AWS often tests the misconception that overfitting is solved by increasing model complexity or data augmentation, but the correct approach is to reduce capacity or add regularization.

How to eliminate wrong answers

Option A is wrong because adding dropout layers is a regularization technique that could help, but it is not the only or most direct action; the question asks for a single action, and reducing epochs or adding regularization (Option B) is a more fundamental fix for overfitting. Option C is wrong because increasing the learning rate can cause the model to diverge or overshoot minima, worsening generalization and potentially increasing overfitting. Option D is wrong because using a different foundation model does not address the root cause of overfitting; the current model is capable of learning the training data, and the issue is with training dynamics, not model architecture.

Full explanation →
476
MCQhard

A machine learning team is building a credit risk model and discovers that the training data has a significant imbalance in loan approval rates between two demographic groups. They decide to reweight the training samples using a preprocessing technique. Which SageMaker Clarify feature can help compute the appropriate sample weights to achieve demographic parity?

A.Clarify preprocessing (reweighting)
B.Clarify post-training bias metrics
C.Model Monitor bias drift
D.Clarify explainability (SHAP)
AnswerA

Clarify provides a preprocessing transformation that reweights data to meet fairness constraints.

Why this answer

Option A is correct because SageMaker Clarify's preprocessing transforms include a reweighting method that assigns weights to instances to adjust for fairness. Post-training (B, C) are not preprocessing. Monitoring (D) is post-deployment.

Full explanation →
477
MCQeasy

A company uses Amazon Rekognition to analyze images for a user-generated content platform. To comply with data residency requirements, they must ensure that images are not processed outside a specific AWS region. What should the company do?

A.Enable Amazon Rekognition VPC endpoints.
B.Use an AWS Lambda function to copy images to a bucket in the desired region and invoke Rekognition there.
C.Configure an AWS WAF rule to block cross-region processing.
D.Set up AWS Organizations to restrict the Rekognition service to one region.
AnswerB

This ensures data is processed only in the specified region.

Why this answer

Option B is correct because it directly addresses the data residency requirement by using a Lambda function to copy images to an S3 bucket in the desired region and then invoking Rekognition in that same region. This ensures that image data never leaves the specified AWS region, as Rekognition processes images only in the region where the API call is made. The other options either do not prevent cross-region processing or are not applicable to Rekognition's data handling.

Exam trap

The trap here is that candidates often confuse VPC endpoints or WAF with data residency controls, not realizing that Rekognition processes data in the region of the API call, and that network-level or application-layer controls do not enforce regional data processing boundaries.

How to eliminate wrong answers

Option A is wrong because Amazon Rekognition VPC endpoints allow private connectivity to the Rekognition API within a VPC, but they do not restrict the region where processing occurs; the API call still goes to the Region endpoint you specify, and data can be processed in any region you choose. Option C is wrong because AWS WAF is a web application firewall that operates at the application layer (HTTP/HTTPS) and cannot control or block cross-region processing of Rekognition API calls, which are made via the AWS API or SDK, not through HTTP requests filtered by WAF. Option D is wrong because AWS Organizations can manage accounts and service control policies (SCPs) to restrict service usage, but SCPs cannot restrict Rekognition to a single region; they can only deny access to the Rekognition API in specific regions, but the company's requirement is to ensure images are not processed outside a specific region, which is a data residency concern, not an API access control issue.

Full explanation →
478
MCQhard

A healthcare company is deploying a model to predict patient readmission risk using Amazon SageMaker. The model processes protected health information (PHI). The company must adhere to HIPAA regulations, which require that all PHI is encrypted at rest and in transit, and that access to the data is logged and audited. The data is stored in an S3 bucket encrypted with SSE-S3. The SageMaker training job uses a custom Docker container that reads data from the S3 bucket and writes model artifacts back to another S3 bucket. The security team notices that the model artifacts in the output bucket are not encrypted. Also, the training logs in CloudWatch Logs contain PHI because the training script logs patient features. The company must fix these issues without delaying the deployment. What should the company do?

A.Implement client-side encryption for the output bucket, and update the training script to mask PHI before logging.
B.Change the output S3 bucket to use SSE-KMS encryption, and enable server access logs for the bucket.
C.Modify the training container to avoid logging PHI, and configure the output S3 bucket to use SSE-KMS encryption.
D.Configure the training job to use a VPC with private subnets, and enable data encryption in transit using TLS.
AnswerC

This directly addresses both issues: encryption of output artifacts and removal of PHI from logs.

Why this answer

Option C is correct because it directly addresses both issues: modifying the training container to avoid logging PHI ensures CloudWatch Logs no longer contain protected health information, and configuring the output S3 bucket to use SSE-KMS encryption ensures model artifacts are encrypted at rest. SSE-KMS is preferred over SSE-S3 for HIPAA compliance because it provides additional controls like key rotation and audit trails. This approach fixes the problems without delaying deployment, as it requires only code and configuration changes.

Exam trap

The trap here is that candidates may focus on network-level controls (VPC, TLS) or logging mechanisms, overlooking the fact that the core issues are encryption at rest for the output bucket and preventing PHI from being written to CloudWatch Logs, which require application-level changes and bucket configuration.

How to eliminate wrong answers

Option A is wrong because client-side encryption for the output bucket would require the training container to encrypt data before writing, which does not address the PHI in CloudWatch Logs; also, client-side encryption is not a native S3 encryption method and adds complexity. Option B is wrong because enabling server access logs for the bucket only logs access requests, not the content of the logs, and does not prevent PHI from being logged in CloudWatch Logs; it also does not address the encryption of model artifacts. Option D is wrong because configuring a VPC with private subnets and TLS encryption addresses data in transit but does not solve the encryption-at-rest issue for the output bucket or the PHI in CloudWatch Logs.

Full explanation →
479
MCQmedium

A financial services company is deploying a generative AI chatbot to assist customers with account inquiries. The company wants to ensure the chatbot does not generate biased or harmful responses. Which combination of AWS services and practices should the company implement to monitor and mitigate these risks?

A.Configure the chatbot to use a pre-trained model from SageMaker JumpStart and disable logging to avoid storing sensitive customer data.
B.Use Amazon Rekognition to analyze chat logs for biased language and automatically block responses with a confidence score above 90%.
C.Use Amazon SageMaker Clarify to detect bias in model outputs and implement a human-in-the-loop workflow with Amazon A2I to review flagged responses.
D.Deploy Amazon Lex with built-in sentiment analysis to detect negative customer emotions and automatically escalate to a human agent.
AnswerC

SageMaker Clarify provides bias detection, and Amazon A2I enables human review of model outputs to ensure responsible AI.

Why this answer

Option C is correct because Amazon SageMaker Clarify is specifically designed to detect bias in machine learning models and their outputs, while Amazon Augmented AI (A2I) enables a human-in-the-loop workflow to review flagged responses. This combination directly addresses the requirement to monitor and mitigate biased or harmful responses from a generative AI chatbot, ensuring responsible AI practices.

Exam trap

Cisco often tests the distinction between services that detect customer sentiment (like Amazon Comprehend or Lex sentiment analysis) versus services that detect bias in model outputs (like SageMaker Clarify), leading candidates to mistakenly choose sentiment analysis options for bias detection.

How to eliminate wrong answers

Option A is wrong because disabling logging prevents the monitoring and auditing necessary to detect biased or harmful responses, and using a pre-trained model from SageMaker JumpStart without additional safeguards does not mitigate bias. Option B is wrong because Amazon Rekognition is an image and video analysis service, not designed for analyzing text chat logs for biased language; it cannot process text-based conversations. Option D is wrong because Amazon Lex's built-in sentiment analysis detects customer emotions but does not detect bias or harmful content in the chatbot's responses, and escalation to a human agent does not proactively mitigate biased outputs.

Full explanation →
480
Multi-Selectmedium

Which TWO strategies can help reduce inference costs when using Amazon Bedrock? (Select TWO.)

Select 2 answers
A.Use a higher temperature setting to generate fewer tokens
B.Increase the max tokens to allow longer responses
C.Use provisioned throughput for high-volume, predictable workloads
D.Cache frequently used responses in Amazon ElastiCache
E.Select a smaller foundation model variant
AnswersC, E

Provisioned throughput offers a discounted hourly rate compared to on-demand per-request pricing.

Why this answer

Using provisioned throughput for predictable workloads reduces per-request cost. Choosing a smaller model variant requires less compute. Caching responses is not directly supported, and increasing max tokens increases cost.

Full explanation →
481
Multi-Selecthard

Which TWO of the following are best practices for preparing training data for a machine learning model?

Select 2 answers
A.Handle missing values by imputing or removing them.
B.Split the data into training, validation, and test sets.
C.Remove all outliers to improve model robustness.
D.Use the entire dataset for training to maximize data usage.
E.Avoid shuffling the data to preserve original order.
AnswersA, B

Missing values can cause errors or bias; imputation or removal is a standard practice.

Why this answer

Option A is correct because handling missing values is a critical data preprocessing step. Missing data can introduce bias or cause algorithms to fail. Imputation (e.g., using mean, median, or model-based methods) or removal of rows/columns with missing values ensures the dataset is complete and suitable for training, preventing errors during model fitting.

Exam trap

Cisco often tests the misconception that removing all outliers is always beneficial, when in fact domain knowledge is required to distinguish between noise and legitimate extreme values that may be critical for model accuracy.

Full explanation →
482
MCQeasy

Refer to the exhibit. A developer runs this command but gets an error: 'An error occurred (AccessDeniedException) when calling the ListFoundationModels operation'. What is the most likely cause?

A.The IAM role does not have bedrock:ListFoundationModels permission
B.The AWS CLI version is outdated
C.The foundation model is not available in us-west-2
D.The region us-west-2 does not support Bedrock
AnswerA

AccessDeniedException is due to missing IAM permissions.

Why this answer

The error 'AccessDeniedException' when calling ListFoundationModels indicates that the IAM role or user executing the AWS CLI command lacks the required permission to list foundation models in Amazon Bedrock. The specific permission needed is bedrock:ListFoundationModels, which must be attached to the IAM identity via a policy. Without this permission, the API call is denied regardless of other factors like region or CLI version.

Exam trap

AWS often tests the distinction between service availability errors (e.g., region not supported) and IAM permission errors, where candidates mistakenly attribute an AccessDeniedException to regional or model availability issues rather than missing IAM permissions.

How to eliminate wrong answers

Option B is wrong because an outdated AWS CLI version would typically produce a different error (e.g., 'InvalidClientTokenId' or 'UnrecognizedClientException'), not an AccessDeniedException, and the ListFoundationModels API is available in recent CLI versions. Option C is wrong because the error is an access denial, not a model availability issue; if a model were unavailable, the error would be something like 'ValidationException' or 'ResourceNotFoundException' when trying to use that specific model. Option D is wrong because us-west-2 (Oregon) fully supports Amazon Bedrock and its APIs; the error is explicitly an IAM permissions issue, not a regional unsupported service error.

Full explanation →
483
Multi-Selecthard

Refer to the exhibit. A data scientist trained an XGBoost model using Amazon SageMaker. Which TWO actions should the data scientist take to improve the model's performance based on the exhibited training job metrics and resource configuration?

Select 2 answers
A.Reduce the VolumeSizeInGB to save costs.
B.Decrease the learning rate and increase the number of training rounds.
C.Increase the maximum number of leaf nodes in the XGBoost algorithm.
D.Use a distributed training strategy by increasing InstanceCount to 4.
E.Switch to a more powerful instance type to reduce training time.
AnswersB, C

A lower learning rate with more rounds typically improves convergence and model performance.

Why this answer

Option A is correct because increasing the maximum number of leaf nodes allows the model to capture more complex patterns, potentially improving AUC. Option C is correct because a lower learning rate with more training rounds often leads to better convergence and performance. Option B (distributed training) primarily reduces training time, not model performance.

Option D (switching instance type) also speeds up training but doesn't directly improve metrics. Option E (reducing volume size) saves cost but does not enhance performance.

Full explanation →
484
Multi-Selecteasy

A company wants to use AWS services to process natural language text. Which TWO AWS services provide natural language processing (NLP) capabilities? (Select TWO.)

Select 2 answers
A.Amazon Translate
B.Amazon Rekognition
C.Amazon Comprehend
D.Amazon Polly
E.Amazon Lex
AnswersC, E

Comprehend provides NLP features like sentiment analysis, entity recognition, and key phrase extraction.

Why this answer

Amazon Comprehend is a natural language processing (NLP) service that uses machine learning to extract insights such as entities, key phrases, sentiment, and syntax from text. Amazon Lex provides NLP capabilities through automatic speech recognition (ASR) and natural language understanding (NLU) to build conversational interfaces like chatbots.

Exam trap

The trap here is that candidates often confuse text-to-speech (Polly) or translation (Translate) with NLP, but these services do not perform language understanding or analysis—they only convert or generate speech/translation without extracting meaning.

Full explanation →
485
Multi-Selecthard

Which THREE considerations are essential for ensuring responsible AI in a model that predicts employee performance? (Choose 3)

Select 3 answers
A.Minimize the number of features to reduce cost
B.Publish the model's predictions publicly for transparency
C.Incorporate human review before final decisions
D.Ensure employee data privacy and consent
E.Test for bias across demographic groups
AnswersC, D, E

Human-in-the-loop reduces automation bias.

Why this answer

Option C is correct because responsible AI frameworks, such as those outlined by AWS, mandate human-in-the-loop (HITL) oversight for high-stakes decisions like employee performance predictions. This ensures that automated outputs are reviewed for context, fairness, and potential errors before affecting employment outcomes, aligning with the AIF-C01 domain's emphasis on human accountability.

Exam trap

Cisco often tests the misconception that transparency means public disclosure of all model outputs, whereas in responsible AI, transparency refers to explainability and auditability of the model's logic, not exposing sensitive predictions.

Full explanation →
486
Multi-Selecthard

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)

Select 3 answers
A.Use a larger instance type with more GPUs.
B.Apply parameter-efficient fine-tuning (PEFT) techniques like LoRA.
C.Increase the batch size to the maximum that fits in GPU memory.
D.Use managed spot training with checkpointing.
E.Enable mixed precision training (FP16).
AnswersB, D, E

LoRA fine-tunes a small subset of parameters, reducing compute and memory.

Why this answer

Option B is correct because Parameter-Efficient Fine-Tuning (PEFT) techniques like LoRA (Low-Rank Adaptation) freeze the pre-trained model weights and inject trainable rank decomposition matrices into specific layers. This drastically reduces the number of trainable parameters (often by 10,000x), lowering memory and compute requirements while preserving model quality, making it ideal for cost- and time-sensitive fine-tuning.

Exam trap

Cisco often tests the misconception that simply scaling up hardware (larger instances) or maximizing batch size is the best optimization strategy, when in fact algorithmic efficiency (PEFT, mixed precision) and cost-saving infrastructure (spot instances) are the correct approaches for balancing cost, time, and quality.

Full explanation →
487
MCQmedium

A machine learning engineer is using Amazon SageMaker to train a model and wants to automatically stop the training job if the loss does not improve for 10 consecutive epochs. Which SageMaker feature should be used?

A.SageMaker built-in algorithms with early stopping
B.SageMaker Training Compiler
C.SageMaker Debugger
D.SageMaker Experiments
AnswerA

Built-in algorithms support early stopping parameters.

Why this answer

Amazon SageMaker built-in algorithms support early stopping, which allows you to automatically terminate a training job when a specified metric, such as loss, stops improving for a defined number of consecutive epochs. This feature is configured directly in the algorithm's hyperparameters (e.g., `early_stopping_patience` for the XGBoost algorithm) and helps save compute time and cost by preventing overfitting.

Exam trap

Cisco often tests the distinction between monitoring (Debugger) and automated action (early stopping), leading candidates to mistakenly choose Debugger because it can detect stagnation, but it cannot stop the job without custom code or a separate hook.

How to eliminate wrong answers

Option B is wrong because SageMaker Training Compiler is a feature that accelerates training by optimizing the model graph and memory usage, but it does not provide any mechanism to monitor metrics or stop training based on convergence. Option C is wrong because SageMaker Debugger is used to monitor training in real time, capture tensors, and set rules (e.g., vanishing gradient detection), but it does not automatically stop the training job; it only emits alerts or saves data for analysis. Option D is wrong because SageMaker Experiments is a tool for organizing, tracking, and comparing training runs, but it has no built-in capability to halt a training job based on metric stagnation.

Full explanation →
488
MCQhard

A financial services company is deploying a fraud detection model using Amazon SageMaker. They need to ensure that inference requests are encrypted in transit and that the model endpoint is only accessible from within a VPC. What is the MOST secure way to achieve this?

A.Enable public endpoint with TLS and restrict access using a VPN
B.Use SageMaker Model Monitor to detect anomalies in request data
C.Create a SageMaker endpoint in a private subnet without an internet gateway, and use AWS PrivateLink
D.Deploy the model endpoint in a VPC with an interface VPC endpoint (AWS PrivateLink) and configure the model to require HTTPS
AnswerD

This ensures traffic stays within the VPC and uses TLS encryption in transit.

Why this answer

Option D is correct because it combines two critical security controls: deploying the SageMaker endpoint within a VPC using an interface VPC endpoint (AWS PrivateLink) ensures that all traffic stays within the AWS network and never traverses the public internet, while configuring the model to require HTTPS enforces encryption in transit. This architecture eliminates exposure to the internet and meets the requirement for encrypted inference requests and VPC-only access.

Exam trap

The trap here is that candidates often confuse 'deploying in a private subnet' with 'making the endpoint accessible within the VPC'—a SageMaker endpoint in a private subnet without an internet gateway or VPC endpoint cannot be reached because the SageMaker service itself needs a network path to invoke the endpoint, and without a VPC endpoint, the endpoint is effectively isolated and unreachable.

How to eliminate wrong answers

Option A is wrong because enabling a public endpoint with TLS still exposes the endpoint to the internet, and VPN access does not restrict access to within a VPC—it only adds an encrypted tunnel from a remote network. Option B is wrong because SageMaker Model Monitor is a feature for detecting data drift and model quality issues, not for controlling network access or encrypting inference traffic. Option C is wrong because creating a SageMaker endpoint in a private subnet without an internet gateway prevents the endpoint from being reachable at all—SageMaker endpoints require network connectivity to the SageMaker service for invocation, and without an internet gateway or VPC endpoint, the endpoint cannot receive requests.

Full explanation →
489
MCQeasy

Refer to the exhibit. A developer wants to choose a model that can generate text (not just embeddings) and has the lowest cost. Based on the exhibit, which model should they select?

A.Titan Embed Text
B.Titan Text Express
C.Titan Text Lite
D.Need more information
AnswerC

Titan Text Lite is a text generation model and is the most cost-effective option among those listed.

Why this answer

Option A, Titan Text Lite, is a text generation model and is the lighter, cheaper option compared to Express. Titan Text Express is more expensive. Titan Embed Text is for embeddings, not text generation.

Therefore, Titan Text Lite is correct.

Full explanation →
490
MCQhard

A healthcare startup uses Amazon SageMaker to train a model predicting patient readmission. They need to ensure the model's predictions do not discriminate based on protected attributes like age or race. Which SageMaker feature allows them to monitor and mitigate bias during training?

A.SageMaker Model Monitor
B.SageMaker Autopilot
C.SageMaker Debugger
D.SageMaker Clarify
AnswerD

Provides bias detection and explainability.

Why this answer

Option B is correct: SageMaker Clarify provides bias detection and mitigation. Option A is wrong: Model Monitor is for data drift. Option C is wrong: Debugger is for debugging training issues.

Option D is wrong: Autopilot automates model building.

Full explanation →
491
MCQeasy

A company wants to use a pre-trained foundation model for sentiment analysis without any customization. Which Amazon Machine Learning service provides access to foundation models via API?

A.Amazon Bedrock
B.Amazon Textract
C.Amazon Comprehend
D.Amazon Rekognition
AnswerA

Why this answer

Amazon Bedrock provides a managed API to access foundation models from providers like AI21 Labs, Anthropic, and Amazon. Amazon Rekognition is for images; Textract for document text; Comprehend for natural language processing (but not foundation models per se).

Full explanation →
492
Multi-Selecteasy

A company is using Amazon Bedrock to generate content for a marketing application. The company wants to ensure that the model does not generate content that violates the company's brand guidelines, which prohibit certain keywords and tones. Which TWO features should the company use to enforce these guidelines? (Choose two.)

Select 2 answers
A.Enable Amazon CloudWatch Logs to capture model output and manually review.
B.Create a prompt template that instructs the model to adhere to brand guidelines and avoid prohibited keywords.
C.Configure Amazon Bedrock Guardrails with custom deny topics and content filters.
D.Use AWS IAM policies to restrict the model's output to only approved words.
E.Encrypt the model responses using AWS KMS to prevent unauthorized viewing.
AnswersB, C

Prompt engineering is an effective way to guide model behavior without additional infrastructure.

Why this answer

Option B is correct because prompt engineering allows the company to embed brand guidelines directly into the instruction given to the model, effectively steering the output away from prohibited keywords and tones. Option C is correct because Amazon Bedrock Guardrails provides a managed, policy-based mechanism to define custom deny topics and content filters that can block or mask unwanted content at inference time, enforcing brand guidelines without manual intervention.

Exam trap

The trap here is that candidates often confuse IAM policies with content moderation, mistakenly believing that IAM can restrict model output vocabulary, when in fact IAM only governs API-level permissions and has no awareness of the semantic content of model responses.

Full explanation →
493
Multi-Selectmedium

A company is deploying an AI-based customer service chatbot using Amazon Lex and AWS Lambda. The company must ensure compliance with GDPR by allowing users to request deletion of their personal data collected during conversations. Which TWO actions should the company take? (Choose two.)

Select 2 answers
A.Use the Amazon Lex DeleteTranscript API to delete user conversation transcripts.
B.Apply an IAM policy that denies access to the conversation logs for the user who requested deletion.
C.Delete the corresponding log streams in Amazon CloudWatch Logs that contain the conversation data.
D.Disable encryption of conversation logs to allow manual deletion.
E.Modify the Lex bot's conversation log settings to stop collecting personal data.
AnswersA, C

DeleteTranscript directly removes stored transcripts for a specific user.

Why this answer

Option A is correct because Amazon Lex provides the DeleteTranscript API, which allows the deletion of user conversation transcripts stored by Lex. This directly supports GDPR compliance by enabling the removal of personal data from conversation logs. The API targets the specific transcripts associated with a user, ensuring that their data is permanently deleted.

Exam trap

The trap here is that candidates may think that restricting access (IAM policy) or stopping data collection (modifying settings) is sufficient for GDPR deletion, but GDPR requires actual erasure of existing data, not just access control or future prevention.

Full explanation →
494
MCQmedium

Refer to the exhibit. A company has the S3 bucket policy shown above. The company uses Amazon Rekognition to analyze images in the 'my-images' bucket. An IAM role used by a Lambda function calls Rekognition. Why might Rekognition be unable to read images from the bucket?

A.The Lambda function does not have an IAM role that allows Rekognition to call S3
B.The bucket policy does not grant Rekognition access to the bucket
C.The bucket policy does not include PutObject permission
D.The condition 'aws:SourceAccount' does not match the account of the Rekognition service
AnswerD

The condition ensures that the request originates from the specified account; if not, access is denied.

Why this answer

The bucket policy includes a condition 'aws:SourceAccount' that restricts access to requests originating from a specific AWS account. If the Rekognition service is being called from a different account (e.g., the Lambda function's IAM role is in a different account than the one specified in the condition), Rekognition will be denied read access to the S3 bucket. This condition is often misconfigured, causing cross-account access failures even when the principal is allowed.

Exam trap

Cisco often tests the nuance that a bucket policy can explicitly grant access to a service principal (like Rekognition) but still block requests due to a condition key mismatch, leading candidates to incorrectly assume the policy is missing the permission entirely.

How to eliminate wrong answers

Option A is wrong because the Lambda function's IAM role is used to call Rekognition, not to directly access S3; Rekognition itself needs permission to read from the bucket, which is governed by the bucket policy. Option B is wrong because the bucket policy does grant Rekognition access (the exhibit shows an Allow effect for the Rekognition service principal), but the condition is blocking it. Option C is wrong because PutObject permission is irrelevant to reading images; Rekognition only needs GetObject (read) permission to analyze images, not write access.

Full explanation →
495
MCQmedium

A company is building a chatbot using Amazon Bedrock to answer customer questions about their product catalog. The chatbot should only use information from the company's internal knowledge base and should not generate answers based on the model's pre-training data. Which feature should be enabled?

A.Use prompt engineering to instruct the model to only use the knowledge base
B.Configure a knowledge base with Retrieval Augmented Generation (RAG)
C.Enable model invocation logging to review responses
D.Fine-tune the model on the product catalog data
AnswerB

RAG grounds responses in the provided knowledge base, avoiding use of pre-training data.

Why this answer

Option B is correct because configuring a knowledge base with Retrieval Augmented Generation (RAG) allows the chatbot to retrieve relevant documents from the company's internal knowledge base and use them as context for generating answers. This ensures the model's responses are grounded solely in the provided data, preventing reliance on its pre-training knowledge.

Exam trap

The trap here is that candidates often confuse fine-tuning with RAG, assuming fine-tuning alone can restrict the model to a specific knowledge domain, when in fact fine-tuning does not prevent the model from using its pre-training data and can still produce off-topic responses.

How to eliminate wrong answers

Option A is wrong because prompt engineering alone cannot reliably prevent the model from using its pre-training data; it only provides instructions that the model may still override with its internal knowledge. Option C is wrong because model invocation logging only records responses for auditing and debugging, it does not constrain the model's source of information. Option D is wrong because fine-tuning adapts the model to the product catalog but does not guarantee that the model will ignore its pre-training data; it can still generate answers from its original training corpus.

Full explanation →
496
MCQhard

An enterprise deploys a foundation model on Amazon Bedrock with a knowledge base. Users report that the model is returning outdated information. What is the most likely cause?

A.The model was fine-tuned
B.The model is not the latest version
C.The knowledge base data source is not refreshed
D.The inference parameters are incorrect
AnswerC

If the underlying data source hasn't been updated, the knowledge base contains stale data.

Why this answer

When a knowledge base is attached to a foundation model on Amazon Bedrock, the model retrieves information from the data source to augment its responses. If the data source is not refreshed, the model will return outdated information even if the model itself is current. Option C directly addresses this by identifying the stale data source as the root cause.

Exam trap

The trap here is that candidates may confuse model versioning (Option B) with data freshness, but the question specifically ties the symptom to the knowledge base, making the refresh cycle the critical factor.

How to eliminate wrong answers

Option A is wrong because fine-tuning adjusts the model's weights on a specific dataset, which does not inherently cause outdated information; in fact, fine-tuning could update the model with newer data. Option B is wrong because using an older model version might affect performance or capabilities, but the question specifically states the model is returning outdated information, which points to the knowledge base content, not the model version. Option D is wrong because inference parameters (e.g., temperature, top_p) control randomness and creativity of responses, not the freshness or accuracy of the information retrieved from the knowledge base.

Full explanation →
497
MCQhard

A research team needs to generate high-quality images with Amazon Bedrock that are realistic and consistent with a specific artistic style. Which combination of parameters should they use?

A.Use a CFG (classifier-free guidance) scale and include a style prompt
B.High temperature, low top_p
C.Low temperature, high top_p
D.Increase the number of steps and reduce the number of samples
AnswerA

CFG scale controls how closely the image follows the prompt; a style prompt (e.g., 'in the style of Monet') ensures artistic consistency.

Why this answer

Option D is correct because for image generation models like Stable Diffusion XL, the CFG (classifier-free guidance) scale controls adherence to the prompt, and a style prompt can enforce artistic consistency. Option A and B (temperature, top_p) are for text models. Option C (increasing steps) improves quality but not style consistency.

Full explanation →
498
Multi-Selectmedium

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

Select 3 answers
A.It can only be used for text data.
B.It provides built-in workflows for image classification and object detection.
C.It supports automated data labeling using active learning.
D.It integrates with Amazon SageMaker to use the labeled data for training.
E.It can only use a public workforce from Amazon Mechanical Turk.
AnswersB, C, D

Ground Truth supports these tasks.

Why this answer

Amazon SageMaker Ground Truth provides built-in workflows for common tasks like image classification and object detection, which simplifies the setup of labeling jobs. These pre-built templates handle the UI and data formatting, allowing users to focus on the labeling instructions rather than building the labeling interface from scratch.

Exam trap

AWS often tests the misconception that Ground Truth is limited to text data or only supports public workforces, while in reality it handles multiple data modalities and offers flexible workforce options including private and vendor-managed.

Full explanation →
499
Multi-Selecteasy

Which THREE factors should be considered when selecting a foundation model for a text generation task? (Select three.)

Select 3 answers
A.Context window length
B.Inference latency
C.Model license
D.Number of parameters
E.AWS Region availability
AnswersA, C, D

Determines the maximum input size, critical for long documents or conversations.

Why this answer

Options A (Context window length), B (Number of parameters), and D (Model license) are key selection criteria. Context window affects input length, parameters affect capability, license affects usage rights. Option C (Inference latency) is operational but often considered after selection.

Option E (AWS Region availability) is relevant for deployment but not model selection.

Full explanation →
500
MCQeasy

Which of the following is NOT one of the core principles of responsible AI as defined by AWS?

A.Transparency
B.Fairness
C.Profitability
D.Robustness
AnswerC

Profitability is not an ethical or responsible AI principle.

Why this answer

The core principles typically include fairness, robustness, explainability, privacy, transparency, and accountability. Profitability is a business objective, not an ethical principle of responsible AI.

Full explanation →
Page 6

Page 7 of 7

All pages

1234567

Practice AIF-C01 by domain

Target a specific domain to shore up weak areas.

Applications of Foundation ModelsFundamentals of AI and MLFundamentals of Generative AIGuidelines for Responsible AISecurity, Compliance and Governance for AI Solutions
See all domains with question counts →