Struggling with Azure AI-102 exam prep? Learn why Storage SAS encryption (C) and built-in term lists (E) aren’t Azure AI Content Safety features for social media apps analyzing English/Spanish/Portuguese content. Boost your certification readiness today!
Table of Contents
Question
Your organization, Xerigon Corporation, is building a social media application using Azure AI Services. The application is expected to be heavily used in both North and South America. Due to regulatory requirements, you want to ensure that the content posted by people is analyzed for the presence of profane language in either English, Spanish, or Portuguese. You use Azure AI Content Safety as the recommended solution for investigating and classifying content in social-media applications.
Which features are NOT available from this service for the social-media application? (Choose all that apply.)
A. Scans text for user input attacks on a large language model
B. Checks whether text responses of large language models use source materials provided by users
C. Uses Storage Access Signatures (SAS) for encryption of data at rest
D. Classifies personal data
E. Uses a built-in term list feature
F. Analyzes images for self-harm content
G. Analyzes text for sexual content
H. Checks AI-generated text for protected material such as song lyrics and recipes
Answer
C. Uses Storage Access Signatures (SAS) for encryption of data at rest
E. Uses a built-in term list feature
Explanation
Azure AI Content Safety does not use the following:
- A built-in term list
- Storage Access Signatures (SAS) for encryption of data at rest
Azure AI Content Safety provides the following:
- Classifies personal data
- Analyzes text for sexual content
- Analyzes images for self-harm content
- Scans text for user input attacks on a large language model
- Checks whether text responses of large language models use source materials provided by users
- Checks AI-generated text for protected material such as song lyrics and recipes
Azure AI Content Safety is the Microsoft-recommended solution for analyzing and classifying content in social-media applications. It provides advanced capabilities to detect, classify, and mitigate harmful and inappropriate content. It can provide classification in different categories such as personal data, auto-corrected text, original text, and language. It leverages the latest AI models to ensure compliance with regulatory requirements, making it the ideal choice for sophisticated and accurate content moderation. It supports multiple languages, including English, Spanish, Portuguese, Italian, and Chinese.
Azure AI Content Safety looks for text that is at risk of user input attacks on a large language model. It provides groundedness detection, which checks if text responses in large language models are grounded in source materials. It also scans for copyrighted material in AI-generated text. It can analyze both text and images for self-harm, hate, violence, and sexual content with different severity levels.
Azure AI Content Safety does not use a built-in or custom term list feature. It uses advanced language and vision models to find content deemed harmful. You can create a custom term list if needed. The Azure Content Moderator service, which has been replaced by Azure AI Content Safety, used a built-in and custom term list feature.
Azure AI Content Safety does not use Storage Access Signatures (SAS) to encrypt data at rest. Instead, it uses customer-managed keys (CMKs) to encrypt and decrypt data at rest. SAS is a URI that can restrict access to Azure blobs, queues, tables, and files.
Microsoft Azure AI Engineer Associate AI-102 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Engineer Associate AI-102 exam and earn Microsoft Azure AI Engineer Associate AI-102 certification.