Azure AI is a platform designed for building and safeguarding generative AI applications. It provides tools and resources to implement Responsible AI practices, allowing users to create, evaluate, and monitor safety systems for their applications.

Azure AI Content Safety monitors text for harmful content such as violence, hate, and self-harm. It allows customization of blocklists and severity thresholds, and it also monitors images for safety. Advanced features include Prompt Shields for detecting prompt injection attacks and Groundedness Detection for identifying ungrounded outputs.

Before deploying your application, use Azure AI Studio’s automated evaluations to test for vulnerabilities and the potential to generate harmful content. The evaluations provide severity scores and explanations to help identify and mitigate risks effectively.