Skip to main content Link Menu Expand (external link) Document Search Copy Copied

Task 2: Set the policies and thresholds for detecting harmful content

  1. On the ContentSafetyXXXX blade, in the navigation pane, select Overview.

  2. Scroll-down and under Make an API call, select Content Safety Studio.

  3. The Content Safety Studio will open in a new tab, in the upper-right corner, select Sign In.

  4. On the Pick an account window, select your lab admin user account @lab.CloudPortalCredential(User1).Username.

  5. Back in Content Safety Studio, select Moderate text content.

  6. On the Moderate text content blade, under Try it Out, verify the checkbox is selected and then select Choose a different resource.

  7. On the Settings blade, select ContentSafetyXXXX and then select Use resource.

  8. Back on the Moderate text content blade, under 1. Select a sample or type your own, select Violent content with misspelling.

  9. Under the 2. Test text box, select Run test.

  10. Under 3. View results, verify the content has been Blocked, and has been rejected using the Violence category.

  11. Also notice the other categories that were judged to be Allowed.

  12. In the upper-left corner, select Content Safety Studio to return to the main page.

  13. Scroll down and then select Prompt Shields.

  14. On the Prompt shields blade, under 1. Select a sample or type your own, select Document attack content.

  15. Under the 2. Test text box, select Run test.

  16. Under 3. View results, verify the results show that jailbreak attacks were detected in the document.