Protecting Online Communities with Azure AI Content Safety
Explore the essentials of Azure AI Content Safety. This course will teach you how to moderate text and image content, and detect harmful or inappropriate material using advanced filtering and prompt shields.
What you'll learn
Managing user-generated content to ensure it is safe and appropriate is a critical challenge for many online platforms. Ensuring that content does not contain harmful, violent, or inappropriate material is essential for maintaining a safe and welcoming environment for all users.
In this course, Protecting Online Communities with Azure AI Content Safety, you’ll learn to effectively moderate and manage content using Azure AI tools. First, you’ll explore how to create and log in to the AI Content Safety Studio Instance.
Next, you’ll discover how to perform text content moderation, including filtering content based on thresholds for hate, violence, sexual content, and self-harm, as well as screening for specific terms using blocklists.
Finally, you’ll learn how to moderate image content and use prompt shields to detect indirect attacks like jailbreaks and prompt injections.
When you’re finished with this course, you’ll have the skills and knowledge of Azure AI Content Safety needed to ensure your platform’s content is safe, compliant, and welcoming for all users.
Table of contents
- Introducing Azure AI Content Safety 2m
- Demo: Create a Content Safety Instance and Launch Content Safety Studio 3m
- Demo: Understanding Safety Categories and Thresholds 3m
- Demo: Blocking Violent Content 2m
- Demo: Using Block Lists 2m
- Demo: Blocking Hate, Sexual, and Self-harm Content 2m
- Demo: Moderating Image Content 4m
- Demo: Using Prompt Shields to Detect Jailbreak Attacks 3m
- Demo: Protected Material Detection 3m