Skip to main content

Video Walkthrough

The Azure AI Content Safety labs takes about an hour to complete end-to-end. It can take longer if you factor in time for exploration and data entry. You might also run into issues or have questions about what should have happened at a specific point in the lab, to help you debug or understand concepts better.

This video recording of the labs can help. The tables below show you the timestamps for each relevant step, within this video. Just visit the YouTube page and open the description to get pre-linked key moments to jump directly to the relevant segments to review walkthrough steps for reference.

1. Prepare Environment​

  • 00:12:41 | Content Safety overview (PowerPoint)
  • 00:47:46 | Launch GitHub codespace for Lab
  • 00:50:03 | Open VS Code. Login to Azure.
  • 00:53:20 | Create Azure resource Set env variables.

2. Analyze Text and Image​

  • 01:05:02 | Open analyze Text Notebook. Select Kernel, Run All.
  • 01:08:03 | Open analyze Image Notebook. Select Kernel, Run All.

3. Playground for NO OpenAI azure subscriptions​

  • 01:11:39 | Requst Access key.
  • 01:13:12 | Launch OpenAI Playground proxy. Authenticate API key.
  • 01:14:28 | Enter user input prompts
  • 01:16:58 | Add System Prompt rules. Test user input prompts.
  • 01:20:15 | Jailbreaking

4. Playground for Azure OpenAI enabled subscriptions​

  • 01:25:41 | Launch Azure portal. Open Azure OpenAI Playground. (portal.azure.com)
  • 01:28:38 | Enter user input prompts
  • 01:29:46 | Add System Prompt rules. Clear Chat. Test user input prompts

5. Handle Jailbreaks​

  • 01:34:13 | Jailbreaks
  • 01:36:34 | Add Content Safety severity levels. Enable Jailbreak detection.
  • 01:40:29 | Add Jailbreak, content severity levels to GPT-3 deployment.
  • 01:41:45 | Test jailbreak detection.