Question

Your team is preparing a customer-facing chatbot for release in Azure AI Foundry. Compliance requires that, before deployment, you quantify the application's likelihood of producing hateful, violent, sexual, and self-harm content when adversarially prompted. You want an automated evaluation that simulates adversarial user inputs and scores the responses against these harm categories. Which evaluation approach should you configure?

Accepted Answer

Run a risk and safety evaluation in Foundry that uses adversarial simulators to generate hostile prompts and scores responses across the hate/unfairness, violence, sexual, and self-harm content categories

Answer

Configure a groundedness evaluator against a test dataset to detect whether responses are supported by the provided context

Answer

Enable coherence and fluency AI-quality metrics on a batch evaluation run over historical conversation logs

Answer

Add a custom code-based metric that counts profanity keywords in the model's output text