SAA-C03 Q568 · Serverless

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?

Accepted Answer

**Option B** satisfies every requirement:

1. **No ML model training required**: Amazon Rekognition is a **fully managed, pre-trained** AI service. Its `DetectModerationLabels` API can identify unwanted content (nudity, violence, drugs, etc.) out of the box — zero training needed.

2. **Image content moderation**: Rekognition's image moderation API is purpose-built for analyzing **still images** (photos), which is exactly what's being uploaded.

3. **Real-time prevention at upload**: The architecture uses an **AWS Lambda function** triggered via a **Lambda function URL**. When a user uploads a photo, the web application calls the Lambda URL → Lambda invokes Rekognition → if unwanted content is detected, the upload is rejected before it's stored.

4. **Serverless & scalable**: Lambda + Reko

📖 Explanation

✅ Correct Answer

❌ Wrong Answers

📘 Key Concepts

🏗️ Architecture

⚙️ Operations

💰 Cost

🔒 Security

🎯 Exam Tips