On this page:

Short Separator

Types of Generative AI Tools Useful for Higher Education

The following list is comprised of tools used by JHU faculty, but access may be division specific.

Short Separator

Using Generative AI Tools to Design Engaging Course Activities

Below are examples of prompting Generative AI tools to generate course objectives, class activities, and test questions. Remember that each example might benefit from additional modifications to the structured prompt where further details and directions for output are provided. Consider the CREATE framework for developing your own prompts.

Create Learning Objectives Efficiently

PROMPT: Write two versions of SMART-based learning objectives to reflect lower and higher cognitive levels: linear regression.

Lower cognitive level (comprehension): By the end of the lesson, students will be able to do the following:

  • Describe the basic principles and assumptions of linear regression analysis
  • Identify the types of variables used in linear regression analysis
  • Interpret the slope and intercept of a linear regression line
  • Explain the concept of correlation and its relationship to linear regression

Higher cognitive level (evaluation): By the end of the lesson, students will be able to do the following:

  • Critically evaluate the appropriateness of linear regression as a statistical method for analyzing a given dataset
  • Design and conduct a linear regression analysis to test a specific research hypothesis
  • Evaluate the validity and reliability of the results obtained from a linear regression analysis and identify potential sources of bias or error
  • Synthesize and communicate the results of a linear regression analysis in a clear and concise manner, using appropriate visualizations and statistical metrics

Both of these learning objectives are SMART based, which means they are Specific, Measurable, Achievable, Relevant, and Time-Bound. The lower cognitive level objective focuses on comprehension and understanding of the basic principles of linear regression, whereas the higher cognitive level objective requires students to critically evaluate, design, conduct, evaluate, and synthesize the results of a linear regression analysis.

Design Massive Self-Assessment Test Pools

PROMPT: Create three multiple-choice assessment questions about linear regression to align with different cognitive levels.

  1. Knowledge (Remembering): Which of the following statements best defines linear regression analysis?  

a. A statistical technique used to analyze categorical data  
b. A statistical technique used to analyze continuous data  
c. A method for identifying the correlation between two variables  
d. A method for identifying the causal relationship between two variables 

  1. Comprehension: Which of the following is the dependent variable in a linear regression analysis? 

a. The variable that is manipulated by the researcher  
b. The variable that is predicted by the independent variable  
c. The variable that is held constant in the analysis  
d. The variable that is used as a control in the analysis 

  1. Application: Given a dataset of student test scores and the number of hours spent studying, which of the following is the independent variable in a linear regression analysis?  

a. Student test scores  
b. Number of hours spent studying  
c. Grade point average (GPA)  
d. Age of the student 

Design Scenario-Based Activities 

PROMPT: List three real examples of risks posed by shadow banking to the economy.

  1. Chinese P2P lending crisis: In 2018, China experienced a crisis in its peer-to-peer (P2P) lending sector, which is a type of shadow banking. Many P2P lenders promised high returns on investments but were found to be fraudulent or insolvent. This led to a collapse of the sector, with many investors losing their savings. The crisis had a significant impact on the Chinese economy, as many investors had to cut back on spending and businesses had difficulty obtaining financing.
  2. The subprime mortgage crisis: In the mid-2000s, shadow banks in the United States, such as mortgage brokers and lenders, engaged in risky lending practices that led to a surge in subprime mortgages. These mortgages were given to borrowers with poor credit and without sufficient collateral, and many were packaged into complex financial products that were sold to investors around the world. When many of these borrowers defaulted, the value of these financial products plummeted, leading to a global financial crisis in 2008. 
  3. European sovereign debt crisis: During the European sovereign debt crisis, many European banks invested heavily in bonds issued by governments in countries such as Greece and Spain, which had high levels of debt. These banks also engaged in complex financial transactions with these countries, such as currency swaps, which helped hide the true level of debt. When these countries began to default on their debts, the banks were left with large losses, which weakened the European financial system and led to a recession. Many of these transactions were facilitated by shadow banking entities, such as investment funds and special purpose vehicles, which were able to avoid regulatory oversight. 

Design Differentiated Learning Activities

PROMPT: Generate prompts at both the remembering level and analyzing level about the secondary mortgage market, credit agencies, credit default swaps, and all the other ingredients that led to the crisis.

Remembering level prompts:

  • What are credit default swaps and how do they work?
  • What was the role of subprime mortgages in the 2008 financial crisis?

Analyzing level prompts:

  • Analyze the role of credit agencies in the 2008 financial crisis. How did their actions and decisions impact the housing market and financial system as a whole?
  • Evaluate the effectiveness of credit default swaps as a risk management tool. How did the widespread use of these instruments contribute to the financial crisis?

Design Various Formats for Activities

PROMPT:  Find a recent news story (from the last 6 to 12 months) showcasing a novel R&D model used by a pharmaceutical company, such as open innovation, merger, acquisition, or the like (use “Changing R&D Models” to identify strategies). Evaluate the strategy using some of the criteria presented in the interview with our expert.

Discussion prompt AI version 1: (using ChatGPT and DALL∙E 2)

Discussion prompt AI version 1: (using ChatGPT and DALL∙E 2)

In the ever-evolving pharmaceutical industry, companies are continuously exploring new models for Research and Development (R&D) to stay ahead of the competition. This infographic showcases some of the novel R&D models adopted by pharmaceutical companies to boost innovation.

Open Innovation Model:

Pharmaceutical companies are increasingly adopting open innovation models to accelerate drug development. This model involves collaborating with external partners such as academic institutions, research organizations, and start-ups to leverage their expertise and resources. This collaboration helps to reduce costs and time-to-market, leading to faster drug development.

Merger and Acquisition Model:

Mergers and acquisitions (M&A) are another popular R&D model adopted by pharmaceutical companies. This model allows companies to acquire new technology, intellectual property, or products, thereby expanding their R&D capabilities. Moreover, M&A can lead to cost savings and operational efficiencies, which can boost innovation.

Virtual R&D Model:

Virtual R&D is a relatively new model that allows pharmaceutical companies to outsource R&D activities to external partners such as contract research organizations (CROs) and academic institutions. This model can reduce costs, improve speed-to-market, and enable companies to focus on their core competencies.

PROMPT: Evaluate the strategy using some of the criteria presented in the interview with our expert.

Discussion prompt AI version 2: (using ChatGPT, DALL∙E 2, and Descript)

Discussion prompt AI version 2: (using ChatGPT, DALL∙E 2, and Descript)
Short Separator

Using Generative AI Tools to Provide Valuable Assignment Feedback

Combine AI Feedback With Human Feedback

Use ChatGPT to provide feedback on a student’s discussion board post.

Student’s answer (85/100):

Reading this week’s readings, I was surprised because I have fallen into the category of being unaware of the arbitration clause. Granted, I haven’t needed to participate in such a manner. However, it’s beneficial to have such knowledge. Both articles stand from fascinating perspectives of the CFPB study. The CFPB argues that its study supports the idea that most consumers aren’t informed of the arbitration clause. Furthermore, it prohibits effective relief for consumers more than a class action lawsuit would. According to CFPB Director, Richard Condray (2015), class actions provide millions of dollars in redress each year. Interestingly enough. Thomas Donohue explains in his article that millions of dollars are awarded through lawsuits, but not in the consumer’s favor. Consumers averagely recover $32, while the plaintiff’s lawyers take home $1 Million on average (Donohue, 2017).

The structures of these arguments are contradictory. The CFPB expresses that they are anti-arbitration because they want the best for consumers and to protect them. Thomas counteracts that ideology through the expression that anti-arbitration is not beneficial and deserves to be protected. In addition, the House of Representatives ruled against it. Paul Bland (2017) reports that the CFPB’s ruling provides consumers the real gift of relief against big banks, such as Wells Fargo, and predatory payday lenders. The ruling would help millions of Wells Fargo’s customers get their day in court to get justice against their fraudulent acts. 

Although I am currently becoming aware of this issue, I am intrigued to discover which side has the best interest for consumers. Why aren’t financial companies forced to vigorous actions depending on the severity of the situation? Why are consumers awarded so little in lawsuit cases? These are questions I would love answered. 

TA’s comments:

Good job analyzing the reasoning behind the conclusions drawn, as well as questioning the bias on both sides. You have provided a detailed and clear analysis. To take this post further, consider incorporating insights from course readings. Did you identify any fallacies in reasoning (Browne ch. 6)? Good job expanding on the topic of discussion with the responses to your classmates—one area of opportunity could be for you to provide counterarguments to your classmates’ claims. This way, you can enhance the discussion instead of simply agreeing with everyone.

ChatGPT feedback:

Overall, the essay provides a brief summary of the author’s reaction to reading two articles about the arbitration clause. The author expresses surprise at not being aware of this issue and notes the contradictory arguments presented by the Consumer Financial Protection Bureau (CFPB) and Thomas Donohue. The essay also mentions the House of Representatives’ ruling against the CFPB’s anti-arbitration stance and the potential benefits of the ruling for consumers.

The essay could benefit from some editing for clarity and coherence.

  1. Sentence clarity: Some sentences in the essay are unclear or confusing. For example, the following sentence could be rephrased for better clarity: “The CFPB expresses that they are anti-arbitration because they want the best for consumers and to protect them.” A clearer way to phrase this sentence could be: “The CFPB takes an anti-arbitration stance because they believe it is in the best interest of consumers and will protect them from potential harm.”
  2. Coherence and organization: The essay could benefit from better organization and flow. One way to improve coherence would be to use transitional phrases or sentences to connect ideas. For example, the essay could use a transitional sentence to connect the paragraph about Thomas Donohue’s arguments with the paragraph about the House of Representatives’ ruling.
  3. Expansion and analysis: The essay could be expanded upon to provide more analysis and reflection. For example, the author could discuss their own opinions on the arbitration clause and provide specific examples of how it has affected consumers. Additionally, the author could compare and contrast the arguments presented by the CFPB and Thomas Donohue to provide a more in-depth analysis of the issue.
Short Separator

Using Zoom AI Companion to Summarize and Organize Recorded Lectures

AI Companion for Zoom is a tool in Zoom that provides several features. It can generate a summary of the group’s conversation. The tool can organize class recordings into chapter and highlights. In all of these cases, the user should review the output for accuracy and edit as needed. While the feature is available in all JHU accounts, it is not turned on by default. Please note that JHU considers student education records to be Personally Identifiable Information (PII). As such, meeting hosts should not enable any of the AI Companion features for meetings involving student education record data (e.g., discussing student grades in office hours). To learn more about the features of AI Companion for Zoom please visit the University Information Systems (UIS) site. For a more generalized overview of recording guidelines with respect to students and within legal compliance, please see the University Guidelines for Recording Class Meetings.