Jailbreak Risk Detection Quiz

What you'll be asked

1. What is the primary purpose of Jailbreak Risk Detection in Foundry?
2. When configuring a Content Safety policy in Foundry, which action is recommended if a prompt is flagged as a high-risk jailbreak attempt?
3. A user provides a prompt that uses 'role-playing' to convince the AI it is in a 'developer mode' where safety filters are disabled. Which Foundry feature is designed to catch this?
4. Why might a 'False Positive' occur within the Jailbreak Risk Detection system?
5. In a production Foundry environment, how does Jailbreak Detection interact with the model inference pipeline?

Answers and explanations come after you submit.