AI as tutor: How Bing Chat can coach students on research and writing


Generative AI certainly has the capability to help with writing and coding, but its skills as a tutor have enormous potential in helping students narrow ideas for papers, learn more about the research process, clarify questions about organizing and focusing their writing, and ask questions when they are working on their own.

Here’s an example of how that works. It uses Microsoft Copilot (formerly Bing Chat) in creative mode, which draws on GPT-4, the most-powerful version of OpenAI’s large language model. It shows examples of prompts and questions that a student might use in beginning a research paper. This process allows the student to ask questions not only about the research process but about definitions and areas the student is uncertain about.

Note how in this example, Copilot is guiding the student rather than simply providing answers the student can copy.

Context prompt

You are an expert in sociology and American history. Act as a tutor and editor to help me narrow and focus the topic of my paper.

Copilot response

I'm happy to help you with your paper. 😊

What is the general topic or theme of your paper?

Prompt

I'd like to explore social changes in small towns during the 20th century. What are some of the main areas that authors have written about?

Copilot response

That's a very broad and interesting topic. Some of the main areas that authors have written about are:

  • The changing demography of rural and small-town America, including population growth and decline, fertility, mortality, migration, ethnoracial composition, and economic inequality¹.
  • The impact of industrialization and urbanization on rural and small-town America, especially in the Northeast and Midwest, where many people moved to cities for factory jobs or were displaced by mechanization and mass production².
  • The role of social class in American history, and how it shaped the experiences and opportunities of different groups of people in rural and small-town America, such as farmers, workers, immigrants, women, and minorities³.
  • The sources and types of social change in rural and small-town America, such as diffusion, ecological change, technological change, and demographic change, and how they affected various aspects of life, such as culture, politics, religion, education, and health⁴.
  • The comparison and contrast of rural and small-town America with other regions or countries that underwent similar or different social changes in the 20th century, such as England⁵.

These are just some examples of possible subtopics that you could explore in your paper. To narrow and focus your topic further, you should consider:

  • What specific time period or geographic area do you want to focus on?
  • What specific research question or thesis statement do you want to answer or argue?
  • What sources of information or evidence do you have access to or plan to use?
  • What is the purpose or goal of your paper? Who is your intended audience?

I hope this helps you get started. If you have any questions or need more assistance, please let me know. 😊

(1) The Changing Demography of Rural and Small-Town America. https://link.springer.com/article/10.1007/s11113-020-09608-5.

(2) Industrialization and Urbanization in the United States, 1880–1929. https://oxfordre.com/americanhistory/view/10.1093/acrefore/978019932917….

(3) Social class in American history - Wikipedia. https://en.wikipedia.org/wiki/Social_class_in_American_history.

(4) Social change | Definition, Types, Theories, Causes, & Examples. https://www.britannica.com/topic/social-change.

(5) An Introduction to 20th-century England | English Heritage. https://www.english-heritage.org.uk/learn/story-of-england/20th-century/.

Follow-up prompt

I’m not sure how to answer all those questions, but I’m mostly interested in the Midwest in the 1970s, 1980s and 1990s. Does that help?

Copilot response

That helps a little bit, but it's still a very broad topic. You need to narrow it down further by choosing a specific aspect of social change that you want to explore in the Midwest in those decades. For example, you could focus on:

  • How the decline of manufacturing and the rise of service industries affected the economy and society of the Midwest in the 1970s and 1980s.
  • How the farm crisis and the environmental movement influenced the politics and culture of the Midwest in the 1980s and 1990s.
  • How the immigration and migration patterns of different groups of people shaped the diversity and identity of the Midwest in the 1970s, 1980s, and 1990s.

These are just some suggestions. You should choose a topic that interests you and that you can find enough information about. Once you have a more specific topic, you can formulate a research question or a thesis statement that guides your paper. For example, if you choose to focus on the farm crisis and the environmental movement, your research question could be:

How did the farm crisis and the environmental movement affect the attitudes and behaviors of Midwestern farmers toward conservation and sustainability in the 1980s and 1990s?

Or your thesis statement could be:

The farm crisis and the environmental movement in the 1980s and 1990s led to a shift in Midwestern farmers' attitudes and behaviors toward conservation and sustainability, as they adopted more innovative and adaptive practices to cope with economic and ecological challenges.

Do you see how these questions or statements are more focused and specific than your original topic? They also indicate what kind of information or evidence you need to support your argument or answer your question. For example, you could use data from census, surveys, interviews, newspapers, magazines, books, documentaries, etc. to show how farmers' attitudes and behaviors changed over time.

I hope this helps you understand how to narrow and focus your topic. If you need more help, please let me know. 😊

Follow-up prompt

My professor told me I needed to build on the work that previous authors have done. I’m not exactly sure what that means. I like the idea of focusing on the farm crisis. Do you have ideas on how I might approach that?

Copilot response

Building on the work that previous authors have done means that you need to review the existing literature on your topic and identify the main arguments, findings, gaps, and debates that have been made by other scholars. This will help you to situate your own research within the broader context of the field and to show how your paper contributes to the existing knowledge or offers a new perspective.

I did a quick search for some sources that might help you with your literature review on the farm crisis. Here are some of the results I found:

  • The U.S. Farm Financial Crisis of the 1980s by Neil E. Harl¹. This book provides a comprehensive overview of the causes, consequences, and responses to the farm crisis, from both an economic and a legal perspective. It also discusses the role of various actors and institutions, such as the Federal Reserve, Congress, farmers' organizations, and lenders.
  • The Farm Debt Crisis of the 1980s: A Review Essay by Gilbert C. Fite². This essay reviews three books on the farm crisis, focusing on different aspects such as the financial, social, and political dimensions. It also compares the farm crisis to previous periods of agricultural boom and bust in American history.
  • Are We Headed Toward Another Farm Financial Crisis as Severe as the 1980s? by Joe Outlaw et al.³. This chapter examines the similarities and differences between the current conditions in agriculture and those that led to the farm crisis in the 1980s. It also analyzes the potential impacts of various policy scenarios on farm income and debt.
  • Taking a look back at the 1980s farm crisis and it's impacts by Mike McGinnis⁴. This article provides a brief summary of the main events and factors that contributed to the farm crisis, as well as some personal stories from farmers who experienced it.

These are just some examples of sources that you could use for your literature review. You should look for more sources that are relevant to your specific topic and research question. You should also critically evaluate each source for its credibility, accuracy, relevance, and bias. You should summarize and synthesize the main points from each source and show how they relate to your own argument or question.

I hope this helps you understand how to do a literature review. If you need more help, please let me know. 😊

(1) The U.S. Farm Financial Crisis of the 1980s - JSTOR. https://www.jstor.org/stable/3744858.

(2) The Farm Debt Crisis of the 1980s: A Review Essay - University of Iowa. https://pubs.lib.uiowa.edu/annals-of-iowa/article/5283/galley/114117/vi….

(3) Are We Headed Toward Another Farm Financial Crisis as Severe as the 1980s?. https://ageco-basis.tamu.edu/extension/resources/downturn-book/chapters….

(4) Taking a look back at the 1980s farm crisis and it's impacts. https://www.farmprogress.com/marketing/taking-a-look-back-at-the-1980s-….

Follow-up prompt

This is a totally new area to me. Where would I find sources?

Copilot response

There are different ways to find sources for your paper. One way is to use online databases or search engines that specialize in academic or scholarly publications, such as Google Scholar, JSTOR, ProQuest, EBSCOhost, etc. These tools allow you to search for articles, books, dissertations, and other types of sources by keywords, titles, authors, dates, subjects, etc. You can also filter your results by relevance, date, type, language, etc. You can access some of these databases through your university library website or with a subscription.

Another way to find sources is to use the references or citations from the sources that you already have. For example, if you read a book or an article that is relevant to your topic, you can look at the bibliography or the footnotes and see what other sources the author used or cited. You can then try to locate those sources and see if they are useful for your paper. This is a good way to find more specific or specialized sources that may not show up in a general search.

A third way to find sources is to ask for recommendations from your professor, your classmates, or other experts in the field. They may have some suggestions for sources that are relevant, reliable, and recent. They may also have some tips on how to evaluate and use the sources effectively.

These are some of the ways that you can find sources for your paper. You should try to use a variety of sources that cover different aspects of your topic and that provide different perspectives or evidence. You should also try to use sources that are current and up-to-date, unless you are doing historical research. You should avoid using sources that are outdated, unreliable, biased, or irrelevant.

I hope this helps you find sources for your paper. If you need more help, please let me know. 😊

Follow-up prompt

OK, so if I go to the library and look for books on the farm crisis, what sorts of things should I search for?

Copilot response

There are different ways to find sources for your paper. One way is to use online databases or search engines that specialize in academic or scholarly publications, such as Google Scholar, JSTOR, ProQuest, EBSCOhost, etc. These tools allow you to search for articles, books, dissertations, and other types of sources by keywords, titles, authors, dates, subjects, etc. You can also filter your results by relevance, date, type, language, etc. You can access some of these databases through your university library website or with a subscription.

Another way to find sources is to use the references or citations from the sources that you already have. For example, if you read a book or an article that is relevant to your topic, you can look at the bibliography or the footnotes and see what other sources the author used or cited. You can then try to locate those sources and see if they are useful for your paper. This is a good way to find more specific or specialized sources that may not show up in a general search.

A third way to find sources is to ask for recommendations from your professor, your classmates, or other experts in the field. They may have some suggestions for sources that are relevant, reliable, and recent. They may also have some tips on how to evaluate and use the sources effectively.

These are some of the ways that you can find sources for your paper. You should try to use a variety of sources that cover different aspects of your topic and that provide different perspectives or evidence. You should also try to use sources that are current and up-to-date, unless you are doing historical research. You should avoid using sources that are outdated, unreliable, biased, or irrelevant.

I hope this helps you find sources for your paper. If you need more help, please let me know. 😊

Follow-up prompt

How can I tell if a source is good?

Copilot response

To tell if a source is good, you need to evaluate it for its credibility, accuracy, relevance, and bias. Here are some questions that you can ask yourself to do this:

  • Credibility: Who is the author or the publisher of the source? What are their qualifications, credentials, or affiliations? Are they experts or authorities on the topic? Do they have a reputation for producing quality work?
  • Accuracy: Is the information in the source correct, complete, and consistent? Does it provide evidence, data, or references to support its claims? Does it acknowledge different viewpoints or limitations? Does it have any errors, mistakes, or contradictions?
  • Relevance: Is the information in the source related to your topic, research question, or thesis statement? Does it address your specific needs or interests? Does it provide new insights or perspectives? Does it have a clear purpose and scope?
  • Bias: Is the information in the source objective, balanced, and fair? Does it present facts or opinions? Does it have any hidden agendas, assumptions, or values? Does it use any emotional or persuasive language?

These are some of the questions that you can ask yourself to evaluate a source. You should also compare and contrast different sources to see how they agree or disagree with each other. You should use sources that are credible, accurate, relevant, and unbiased.

I hope this helps you tell if a source is good. If you need more help, please let me know. 😊

Follow-up prompt

My professor told me that the best history has a point of view and that primary sources always have a point of view. How can I write with a point of view and still be unbiased? And how can I tell what biases a source might have?

Copilot response

Writing with a point of view and being unbiased are not necessarily contradictory. A point of view is your perspective or interpretation of the topic, based on your research, analysis, and argument. Being unbiased means that you are fair, objective, and open-minded, and that you acknowledge and address different points of view, evidence, and sources.

To write with a point of view and still be unbiased, you should:

  • Have a clear and specific research question or thesis statement that guides your paper and expresses your main argument or answer.
  • Use a variety of sources that are credible, accurate, relevant, and unbiased to support your point of view and to show different perspectives or evidence.
  • Analyze and synthesize the information from your sources and explain how they relate to your point of view and to each other.
  • Acknowledge any limitations, gaps, or uncertainties in your research or argument and suggest ways to overcome or address them.
  • Use clear, precise, and respectful language that avoids emotional or persuasive appeals or generalizations.

To tell what the biases a source might have, you should:

  • Consider the author's or the publisher's background, purpose, audience, and context. What are their qualifications, credentials, affiliations, or interests? What are they trying to achieve or convey? Who are they writing for or to? When and where did they write or publish?
  • Examine the content and the tone of the source. What are the main points or arguments? What are the facts or opinions? What are the assumptions or values? What is the evidence or data? How is the language used?
  • Compare and contrast the source with other sources. How does it agree or disagree with other sources? How does it fit or differ from the general consensus or trend? How does it reflect or challenge the dominant or alternative views?

These are some of the ways that you can write with a point of view and still be unbiased, and that you can tell what the biases a source might have. I hope this helps you with your paper. If you need more help, please let me know. 😊

What now?

This interaction with Copilot (which was known as Bing Chat when this simulation was done) is just one of many possibilities for using generative AI in classes. As you explore other approaches to integrating generative AI into your teaching, you might consider a few questions:

  • What types of interactions with chatbots would help students deepen their understanding of course material or allow them to explore areas we usually don't cover in class?
  • What additional roles might a chatbot take on to help guide students in their studies?
  • What routine tasks could chatbots do that would free up student time for deeper learning?
  • What types of assignments would be most appropriate for use with chatbots?
  • What can I learn from students about how they might benefit from use of generative AI?

You will no doubt have additional questions. That's good. This is an important time for experimentation and finding new ways of helping students learn.

Assigning AI: Seven Ways of Using AI in Class, by Ethan Mollick. One Useful Thing (12 June 2023). Also see Assigning AI: Seven Approaches for Students With Prompts, by Ethan Mollick and Lilach Mollick (11 June 2023), which goes into more detail about each approach.

Brainstorming With AI, by Owen Holt. Association for Talent Development (30 June 2023).

Experimenting with using ChatGPT as a simulation application, by Bryan Alexander. Bryan Alexander blog (5 March 2023).

How to Use AI to Do Stuff: An Opinionated Guide, by Ethan Mollick. One Useful Thing (15 July 2023).

New Modes of Learning Enabled by AI Chatbots: Three Methods and Assignments, by Ethan R. Mollick and Lilach Mollick. SSRN (23 December 2022).

Using Generative AI to Illustrate Concepts, by Eric Steinhart. AutomatED (29 May 2023).

What AI can do with a toolbox … Getting started with Code Interpreter, by Ethan Mollick. One Useful Thing (7 July 2023).