Microsoft Copilot is an AI-powered assistant available to UMB. Copilot says it “can handle your questions and complete tasks for you via generative AI. Copilot can be used to solve complex tasks like math problems, create images using Bing’s Image Creator, and more.” Copilot protects UMB users' data, so it’s important to always sign into your UMB Microsoft account when using it.

For help accessing Copilot, view step-by-step login instructions here.

* Copilot used to be referred to as Bing Chat Enterprise.

While some AI-based tools might immediately come to mind, there are a wide variety of tools available, each serving different purposes. As you evaluate how to use these tools responsibly in your academic work, it is useful to keep the breadth of potential uses in mind. You or your instructor might want to use one kind of tool for a certain situation but not others.

Text and code

Some tools will give you writing or paraphrasing suggestions (e.g. Grammarly, QuillBot), and other tools will generate new text or code based on a prompt from you (e.g. ChatGPT, Microsoft Copilot, Claude, Google's Gemini). We refer to these tools as Generative AI tools. 

Images

Some tools will generate unique images for you, based on a text prompt (e.g. DALL-E, Adobe Firefly, Copilot Designer).

Research

Some tools are designed to help you find research articles or links (e.g. Perplexity, Elicit).

Alongside the exciting possibilities of these AI-based tools, there are a number of things to be careful about as you assess if and when you want to use them. Consider the following:

Copyright and Labor

Where does the content come from? Because machine learning takes huge inputs of data sets, many models use information from the internet in their training. Artists and authors have criticized AI-based tools for using their work without compensation or credit. If an AI-based image generator can produce work in the style of a certain artist, should that be seen as stealing, or paying homage?

Bias

While it may be tempting to think of an output from an AI-based tool as neutral when it comes to bias, that is not the case. Since machine learning models are trained on real-world datasets, and since the world contains bias, it is safe to assume that outputs from these models may replicate or even exacerbate biases we see in the world around us.

Security and Privacy

It is safe to assume that -- in some way or another -- any information you put into an AI-based tool is being used to further train the machine learning model. If you choose to use these tools, you'll want to make sure you're never putting personal or secure information about you or anyone else in your chats. You should also read through any user agreements if you sign up to use a particular service and decide if you are comfortable agreeing to the terms. If one of your class projects requires the use of a particular technology that you do not wish to create an account for, you can ask your instructor for an alternative way to complete the assignment. 

Here are four facts about ChatGPT that will be helpful as you think about its uses.

ChatGPT was developed by a company called OpenAI, based in San Francisco, CA. Their stated mission is "to ensure that artificial general intelligence—AI systems that are generally smarter than humans—benefits all of humanity." ChatGPT is just one of OpenAI's products. OpenAI began as a nonprofit in 2015 and then became a for-profit company in 2019. They received investments from different sources, including $1 billion from Microsoft, which uses the same AI model that powers ChatGPT and DALL-E for Bing AI.  

The "GPT" in ChatGPT stands for Generative Pre-trained Transformer. 

• Generative refers to a type of machine learning model that creates new content.
• Pre-trained means it was trained on an enormous amount of data. In the case of ChatGPT, the model was trained on hundreds of billions of words (mostly from websites) to learn patterns and relationships between words and phrases.
• Transformer refers to part of the machine learning model that can better comprehend sentences because it can recognize the relative importance of different words in context.

It can generate new text for any written prompt and is designed to provide human-like conversational interactions. Some examples of things it can do include: draft an email, generate a definition, de-bug or write functional code, create an outline for writing assignments, etc.

It produces unique responses each time, and you can tweak its responses because it remembers the whole chat conversation. You can instruct it to respond to you in different ways (e.g. "please reply as if I'm a first-year medical student").

Can you trust that the information you receive from AI-based tools is correct? Not without double-checking. Many chatbots such as ChatGPT were designed to produce content that seems realistic, so they will produce inaccurate content with the same level of confidence as accurate content and it's up to you to determine which is which. You'll need to employ a variety of strategies to double-check information before assuming it is correct. You'll learn more about how to do this in the Assess Content section of this module.

Generated with Microsoft Copilot Designer . July 22, 2024.

Prompt: Health Sciences and Human Services Library in Baltimore

If you have a question or comment about this guide, or how the HSHSL can support you, contact Emilie Ludeman, Research and Education Librarian, at eludeman@hshsl.umaryland.edu.