Preventing Unwanted Image Uploads Without ML Training

SAA-C03 By islever Editorial Last updated

Design secure workloads and applications.

Short answer

B Use an AWS Lambda function with Amazon Rekognition to detect unwanted content in uploaded photos without training a custom model.

Question

A company needs a solution to prevent photos with unwanted content from being uploaded to the company's web application. The solution must not involve training a machine learning (ML) model. Which solution will meet these requirements?
  1. Create and deploy a model by using Amazon SageMaker Autopilot. Create a real-time endpoint that the web application invokes when new photos are uploaded.
  2. Create an AWS Lambda function that uses Amazon Rekognition to detect unwanted content. Create a Lambda function URL that the web application invokes when new photos are uploaded. Correct Answer
  3. Create an Amazon CloudFront function that uses Amazon Comprehend to detect unwanted content. Associate the function with the web application.
  4. Create an AWS Lambda function that uses Amazon Rekognition Video to detect unwanted content. Create a Lambda function URL that the web application invokes when new photos are uploaded.

Key Takeaways

The exam tests knowledge of managed AI/ML services, specifically distinguishing between image analysis (Rekognition), text analysis (Comprehend), and custom model creation (SageMaker). The trap is choosing services that don't match the data type or require unnecessary training.

This question tests the correct AWS service for content moderation of images without custom model training. The solution requires using Amazon Rekognition to detect unwanted content via a Lambda function triggered by uploads.

Common mistake: Candidates often choose Amazon SageMaker Autopilot because it sounds like 'AI', but the question explicitly forbids training a model. Others might pick Comprehend, which is for text, not images.

Expert Analysis

Why the Answer Is Correct

Option B is the correct answer because Amazon Rekognition is a fully managed computer vision service that can detect objects, scenes, labels, faces, and explicit content in images out-of-the-box. Since the requirement is to prevent uploads of photos with unwanted content without training a new ML model, Rekognition is the ideal managed service. Using an AWS Lambda function allows the web application to invoke this logic asynchronously or synchronously upon upload, ensuring the check happens before storage.

Why the Other Options Are Wrong

Option A suggests using Amazon SageMaker Autopilot, which automates the building of ML models but still involves creating and deploying a custom model, violating the 'no training' constraint. Option C uses Amazon CloudFront functions with Amazon Comprehend; Comprehend is a natural language processing service for text, not images, making it irrelevant for photo analysis. Option D uses Amazon Rekognition Video, which is designed for analyzing video streams or files, not static images, making Rekognition Image (part of Rekognition) the more precise and appropriate choice for photos.

Community Comment Notes

Community consensus strongly supports Option B, with multiple users noting that Rekognition is the standard service for image and video analysis. One user highlighted the distinction: 'Rekognition: for image and video analysis... Comprehend: natural language processing'. Another user confirmed that Rekognition includes both Image and Video capabilities, reinforcing why Rekognition (general) or Rekognition Image is the right path over SageMaker or Comprehend.

Exam Strategy

Always read the constraints carefully. If the question says 'no training' or 'out-of-the-box', avoid SageMaker and look for managed AI services like Rekognition, Textract, or Polly. Match the data type (image vs. text vs. audio) to the correct service name.

Frequently Asked Questions

Why is SageMaker Autopilot incorrect?

Autopilot creates custom ML models, but the question explicitly forbids training a machine learning model.

Can I use Amazon Comprehend for images?

No, Comprehend is for natural language processing of text, not for analyzing visual content like photos.

More SAA-C03 FAQ →

Community Votes

B
100%

100% of anonymous learners picked answer B. Votes are pick records left by other test-takers — they are not the verified answer.

Discussion 5 contributions

Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded, and comments reflect the views of individual contributors.

From public community threads 5

  1. HarryLopez 👍 6 Answered B
    Rekognition: for image and video analysis Comprehend: natural language processing model for uncovering insights and connections in text Sagemaker Autopilot: feature set that simplifies and accelerates and automates the various stages of the machine learning workflow
  2. LeonSauveterre 👍 4 Answered B
    You've made it this far, and you're all gonna make it. For your quick reference: Vision and Image Processing 1. Amazon Rekognition: Detects objects, scenes, text, faces, and activities in images/videos. 2. Amazon Textract: Extracts text, tables, and forms from scanned documents. Speech and Language 1. Amazon Polly: Converts text to lifelike speech. 2. Amazon Transcribe: Converts speech to text for audio files. 3. Amazon Translate: Yeah, translate. 4. Amazon Comprehend: Extracts insights like sentiment, entities, and key phrases from text. Machine Learning Development 1. Amazon SageMaker: Comprehensive ML platform for building, training, and deploying models. 2. Amazon SageMaker Autopilot: Automatically builds, trains, and tunes ML models, basically simplified model creation for non-experts. 3. Amazon SageMaker JumpStart: Provides pre-built ML models and solutions for quick deployment.
  3. zdi561 👍 1 Answered A
    Amazon Rekognition is a visual analysis service that includes both Amazon Rekognition Image and Amazon Rekognition Video. Amazon Rekognition Image analyzes images, while Amazon Rekognition Video analyzes videos.
  4. Andy_09 👍 8
    Option B
  5. NayeraB 👍 3 Answered B
    B is correct

Reader comments & corrections

No reader corrections yet — spotted an error or have a note? Share it below.

Log in to comment, report an error, or add a note about this question.

Submitted for moderation before publishing. Keep it helpful and respectful.

S3 Multi-Region Access Point and Replication Strategy

S-grade

Securing NLB with Security Groups for Minimal Change

S-grade

Most Cost-Effective 7-Year Retention for EBS Snapshots?

S-grade

AWS Forecast vs SageMaker for Time-Series Prediction

S-grade

Practice All SAA-C03 Questions

Access 120 questions with complete answers and detailed explanations.

View Full SAA-C03 Practice Test →

← Back to SAA-C03 Study Guide