There is an AI that was introduced by Microsoft. This AI is responsible for helping organizations measure, mitigate and manage AI risks.
Phi users can now get Azure AI Content Safety features, including Prompt Shields, Protected Material Detection, and Ground Adhesion Detection. These features can be used as content filters for any language model in our model library, and developers can easily incorporate them into their applications via a single API. Once in production, developers can monitor their application for quality and security, adversarial prompt attacks and data integrity, and take timely action using real-time alerts.