Guide to CWRU AI

CWRU AI is an internal, secure, general purpose AI portal for the CWRU community. When you log into CWRU AI, you are able to select the large language model you want to use to power the chat, currently Open AI’s ChatGPT 4o or 3.5 Turbo, with other models coming soon. To learn more about how to chat effectively with generative AI, check out this video on LinkedIn Learning.

Key features of CWRU AI

  • Regardless of the model you pick, data entered into the portal are not shared outside the university and are never used to train the underlying models.
  • You can upload and query your own files, including images, PDFs, Word, Excel and CSV files. More file types will be added in the future.
  • You can upload multiple files at one time into a chat conversation.
  • Your conversation history, including any files you uploaded to the conversation, is saved so that you can refer back to it.
  • You can rename conversations to make them easier to find.
  • You can maintain a personal prompt library to re-use helpful prompts.
    • A prompt in generative AI is a natural language request that a user enters into a generative AI tool to produce a response. Prompts can be questions, instructions, examples, or contextual information. The quality of the response depends on the quality of the prompt.

Logging in

  1. Browse to https://ai.case.edu.
  2. Click on the "Login with Microsoft" button.
  3. Enter your CWRU email address in the form of abc123@case.edu and click "Next".
  4. Enter your CWRU passphrase and click "Sign in".
  5. If prompted, confirm Duo.
  6. Click the "Guidelines" button, read the information about the bot, and click "I Agree". As guidelines may change over time, you will be prompted to review and agree to the guidelines each time you log in or select a new bot to chat with.

Chatting

  1. In the "type a message" box, type in your question or request. You can also click the microphone icon to speak your inquiry.
  2. Hit "Enter" on your keyboard or the up arrow button to submit your question to the chatbot.

Asking Questions of Uploaded Documents

  1. Click the paperclip button, then the "browse" link.
  2. Browse to and select the file or files you want to include in the chat. You can upload multiple files at the same time.
  3. Click the "Open" button.
  4. Click the "Chat with file" button.
  5. Once the file uploads, you can start asking questions related to the content of the file(s).

Saving and Accessing Prompts

  1. You can create a personal library of useful prompts. To add a prompt to your library for repeated use, click the prompt button (looks like two sheets of paper).
  2. Click "Add Prompt".
  3. Type in a title for the prompt and the prompt text and click "Create Prompt".
  4. To access your prompt library, click the same prompt button.
    • Example: Save a prompt to proofread documents you upload, such as "Proofread this document for grammar and punctuation. Provide back the corrected text as well as a bulleted list of your corrections and changes."

Starting a New Chat and Accessing Previous Chats

  1. Whenever you switch topics with a chatbot, you will get better results if you start a new chat. Click the "New Chat" button in the upper right to start a new conversation.
    • Note: if you do not see a blue menu on the left of the screen with a "New Chat" button at the top, click the three line menu in the upper left corner to open the menu.
  2. Previous conversations are listed underneath the "New Chat" button. You can return to and resume any conversation by clicking its title. If you have uploaded documents to that chat, they will still be present.
  3. If you click the ellipsis button (...) next to a previous chat, you can rename, archive, export, or delete it.
    • Example: If you have a chat with multiple documents uploaded into it, you can rename the chat to something easily recognizable and return to that chat repeatedly so that you do not need to re-upload those documents each time you want to query them.

Changing to a Different GPT

  1. The default GPT when you log into CWRU is "GPT 4o", which uses Open AI’s GPT 4o large language model. Other GPTs with different language models may be available to you.
  2. To access them, click the "New Chat" button and then click the dropdown arrow next to "GPT 4o" to see and select other GPTs that are available to you.
    • Example: different language models have their own strengths, weaknesses, and styles. Try the same prompt across multiple GPTs to see which one gives you the best response for that particular question or task

Installing CWRU AI on Android Devices

  1. Go to ai.case.edu in Chrome and log in
  2. Touch the vertical ellipsis menu in the upper right corner and select "Add to Home screen"
  3. Touch "Install" and then "Install" again
  4. A "nebulaONE" icon will be added to your Android devices home screen

Adding CWRU AI as a Home Screen Shortcut on Apple iOS Devices

  1. Go to ai.case.edu in Safari and log in
  2. Touch the share button at the bottom center of the screen and select "Add to Home screen"
  3. Touch the share button at the bottom center of the screen and select "Add to Home screen"
  4. If prompted, name the shortcut
  5. Touch the "Add" button in the upper right corner
  6. An icon will be added to your Apple devices home screen