Generative AI and Teaching
Resource Overview
Information about GenAI tools and their implications for teaching and learning
What is Generative AI (GenAI)?
Generative AI (GenAI) refers to a host of tools that can generate new content such as text, images, or audio. These tools are trained on large amounts of data and can perform many tasks much faster than could be done manually by a real person. Generative AI composition tools, in particular, can write convincing text on nearly any topic and even have conversations that mimic human interaction. While GenAI itself is not new, it became much more widely used with the public release of new versions of improved composition tools in 2023, such as ChatGPT.
The ability to quickly and easily generate relatively high-quality text and images raises a number of concerns. Faculty at WashU and other institutions have wondered how GenAI will impact teaching and learning. On one hand, GenAI creates new possibilities to support student learning through things like interactive learning tools, self-quizzes, and help with brainstorming, among others. On the other hand, many instructors worry about the integrity of take-home exams and essays, the future utility of homework, and if GenAI will discourage students from learning the fundamentals of a topic if they can simply ask chatbot for an answer. However as GenAI is not going to go away, instructors will likely need to adapt their courses to both make use of GenAI (when appropriate) and to respond to any challenges that AI may present to traditional models of teaching and learning.
Recommendations for GenAI and Teaching
Adapting your courses to the new AI landscape can feel daunting. While there is no single correct approach to do so, we hope that the following steps will help you make the pedagogical decisions that are right for you and your students.
- Learn about GenAI and its capabilities firsthand
- Reflect on the role of GenAI in your courses and discipline
- Consider ways that GenAI may enhance student learning
- Consider ways that GenAI may detract from student learning
- Reflect on the ethics of AI use and avoidance
- Be transparent about the role of AI in your course and clear about your AI policies
1. Learn about GenAI and its capabilities firsthand
The best way to understand GenAI’s capabilities and limits is to try it out yourself. Experiment with some of the major AI tools that generate text and images. Compare the text results to the Washington University versions of ChatGPT Beta and Microsoft Copilot. Learn the basics of effective prompt writing to really understand what GenAI can do. See how GenAI responds to your course assignments. And check back occasionally as AI capabilities and the AI landscape are changing quickly.
2. Reflect on the role of GenAI in your courses and discipline
The impact of generative AI varies across disciplines and across courses within disciplines. Some fields may expect graduates to use AI daily, while other fields may take a much more cautious approach. When confronted with any potentially disruptive situation in teaching and learning, we recommend returning to your students’ learning goals and the principles of backwards design. What should students know or be able to do after taking your course? How will you and your students know if they have achieved those goals? How will you help them reach those goals? Focusing on these kinds of fundamental questions can help you make informed decisions about the appropriate role of GenAI in your course.
3. Consider ways that GenAI may enhance student learning
Students’ progress towards some learning goals may be helped by incorporating GenAI into your teaching and assignments. For example, GenAI tools like ChatGPT can be added into assignments or you can use AI to generate examples and exercises to use in class. Faculty from many institutions have used GenAI in their courses in creative ways. We encourage you to investigate ideas and resources for using AI within your teaching and assignments. One place to start is the book Teaching with AI: A Practical Guide to a New Era of Human Learning which contains many creative ideas for using AI as an instructor (available as an ebook from WashU libraries). Another is WashU’s new Generative AI Teaching Activities: Online Repository.
4. Consider ways that GenAI may detract from student learning
Sometimes students must first learn the basics of a field in order to achieve long-term success, even if they might later use shortcuts when working on more advanced material. We still teach basic mathematics to children, for example, even though as adults we all have access to a calculator on our smartphones. GenAI can also produce false results (aka “hallucinations”) and often only a user who understands the fundamental concepts at play can recognize this when it happens. Finally, sometimes the end product students produce isn’t even the focus in education per se. Many writing assignments, for example, are much more about learning how to think and to make arguments than about simply creating a block of text on a particular topic. Like a runner training for a marathon, the point isn’t to just get from point A to point B, but to transform oneself along the way. A car might be faster, but it won’t build the muscle and endurance we seek. In these cases, it may make sense to adopt some teaching strategies that avoid AI and assignments that are more AI resistant.
In these cases, investigate ideas and resources for AI resistant teaching and assignments.
5. Reflect on the ethics of AI use and avoidance
There are many unresolved ethical questions surrounding GenAI and its use. For example, the results of GenAI often display the biases of their training data, the use of AI may have an environmental impact, there are unresolved questions related to intellectual property and AI models, not everyone has equal access to all AI tools and robust AI literacy, and AI citation practices can be extremely murky. At the same time, some have also argued that we would actually be doing our students a disservice not to train them to use GenAI, especially as it seems some fields may expect our students to use it upon graduation. All of these issues are complex and unlikely to be resolved soon, but are important to consider as you make decisions about the use of AI in your courses.
6. Be transparent about the role of AI in your course and clear about your AI policies
Be explicit in your course about when/how students can or cannot use AI tools and your reasons for making those decisions. Different instructors will have different expectations in regards to GenAI, so being clear about your expectations for your course is critical. For assignments, identify skills that are crucial for your students to practice and those that are less crucial and would benefit from AI support. There is a broad spectrum of ways students and AI could contribute to an assignment. Let your students know what is acceptable in your course by adding a short statement to your syllabus explaining your GenAI course policies.
For more ideas, see these example tables for more potential spectrums of AI-Student Collaboration that you can copy/use in your class. For example, boundaries can vary across assignments and across language classes. At the intro level, students may need to practice grammar, so students must proofread on their own for each assignment. For other classes, proofreading and getting feedback from AI may be acceptable, but writing the core arguments and content may not be. For other course, AI could be used to generate possible ideas that students then curate, build on, revise, or synthesize.
Further Resources
- Basic Information
-
- ChatGPT (Link to the tool)
- What Is ChatGPT? A Basic Explainer (PC Magazine)
- Things to Consider When Using AI (Missouri Online; UMSL)
- WashU Learn AI Collection (Curated collection of LinkedIn Learning modules)
- Teaching Activities
-
- Generative AI Teaching Activities: Online Repository (WashU CTL)
- Teaching with AI: A Practical Guide to a New Era of Human Learning (available as an ebook from WashU libraries)
- AI Pedagogy Project (metaLAB (at) Harvard)
- News Outlets
-
- Anatomy of an AI Essay (Elizabeth Steere, The Chronicle of Higher Education)
- Professors Ask: Are We Just Grading Robots? (Beth McMurtrie, The Chronicle of Higher Education)
- When It Comes to Critical Thinking, AI Flunks the Test (Gary Smith and Jeffrey Funk, The Chronicle of Higher Education)
- What Will Determine AI’s Impact on College Teaching? 5 Signs to Watch (Beth McMurtrie, The Chronicle of Higher Education)
- Instructors Rush to Do ‘Assignment Makeovers’ to Respond to ChatGPT (Jeffrey Young, EdSurge)
- EdSurge reporting on AI
- Blog Posts
-
- Three Recent Studies on Student Learning with Generative AI (Derek Bruff, University of Mississippi)
- Simulating History with Multimodal AI: an Update (Benjamin Breen, UCSC)
- Should You Add an AI Policy to Your Syllabus? (Kevin Gannon, Queens University of Charlotte)
- Let ChatGPT Be Your Teaching Assistant (Ethan Mollick and Lilach Mollick)
- ChatGPT Educational Friend or Foe (Kathy Hirsh-Pasek and Elias Blinkoff, Temple University)
- Additional Resources
-
- How Generative AI is Changing the Classroom (Graham Vyse, The Chronicle of Higher Education with Amazon Web Services)
- Prompt Library: Instructor Aids, Student Exercises, and Other (Ethan Mollick and Lilach Mollick, University of Pennsylvania)
- ChatGPT Assignments to Use in Your Classroom Today (pdf, Yee et al, University of Central Florida)
- Practical AI for Teachers and Students (Ethan Mollick and Lilach Mollick, University of Pennsylvania)
- Assigning AI: Seven Approaches for Students, with Prompts (Ethan Mollick and Lilach Mollick, University of Pennsylvania)
- Artificial Intelligence and the Future of Teaching and Learning (US Department of Education)