The content filtering system is powered by Azure AI Content Safety, and it works by running both the prompt input and completion output through an ensemble of classification models aimed at detecting and preventing the output of harmful content.
In this demo, we’ll demonstrate how to create and configure a content filter.
Disclosure: This demo contains an AI-generated voice. This video also contains harmful content for demonstrative purposes.
Chapters:
00:00 – Introduction
01:40 – Create a content filter
02:16 – Create a blocklist
03:45 – Test the content filter
Resources:
Azure AI Studio – https://www.ai.azure.com
Responsible AI Developer Resources – https://aka.ms/operationalize-rai
To learn more about Harm Categories and Severity Levels – https://www.aka.ms/rai-harm-categories