Artificially Intelligent tools have been in the background of our lives for many years now, as search engines,
virtual assistants, language translators, facial recognition softwares, stock trading applications, tutoring tools, grading, etc.
None of these previous iterations have generated the amount of interest and reaction as the ChatGPT (Chat Generative Pre-trained Transformer)(OpenAI) has.
ChatGPT scans the texts of the Internet and returns answers to questions that are detailed and articulate, though often inaccurate or superficial. Since its release in late November 2022, it has been the subject of reactions from multiple
perspectives. New York City banned ChatGPT use in schools, as did Seattle, WA. ChatGPT is feared to be able
to replace CEOs and writers of all kinds. Media is abuzz with ChatGPT and so are faculty in Higher Education.
We at OTEAR have been following the discussions of ChatGPT and aim to provide our colleagues with useful and
helpful information for their consideration in their practice. The resources we provide here cover the basics of what ChatGPT is and how it works, as well as provide ideas for ways it can be used in the classroom.
AI Round Table
Announcing the AI Round Table, a space for the Rutgers community to come together to learn about and discuss issues related to AI in Higher Education. Events held in this space will include discussions about AI and pedagogy, academic integrity and generative language models, and professional use of AI as part academic and administrative work.
Special Events from the AI Round Table
AI Tools for Teaching and Learning – A Petting Zoo (Virtual Interactive Workshop)
Monday, June 12, 2023, 11:00am-12:30pm We've been discussing AI concerns and issues. Let's also discuss promising strategies for the productive, creative, and ethical use of AI in our classrooms and with our students. Join us and special guest Associate Professor Joyce Valenza for an AI petting zoo, where we will explore and test-drive applications for productivity, research, and instructional design and consider the framing of new norms for integration the coming semester (You can learn more about Joyce’s work here). If you wish to attend the workshop or to receive the video recording, please register here.
Suggestions for Rutgers Faculty
1) Learn what AI tools can and cannot do by reading up on these tools and experimenting with
them before incorporating an AI tool into a class activity or restricting its use. Become
familiar with the student conduct requirements for charging students with violations of academic
integrity if your choice is to restrict its use. Resources available on the OTEAR Canvas site.
2) Consider developing assignments that require students to use higher order thinking, connect
concepts to specific personal experiences, cite class readings and discussions, and make innovative
connections. These types of prompts are more difficult for students to answer using AI tools. Ideas
about assignment design are also available on the OTEAR Canvas site.
3) Develop rubrics that emphasize critical thinking, application of knowledge, and evaluation of
knowledge rather than the restatement of material or summary. A sample rubric is given on the OTEAR
Canvas site.
4) Develop clear policies for each course. For example:
“Use of AI such as ChatGPT is not permitted in any stages of the writing process on any assignment.”
“Use of AI such as ChatGPT is only permitted to help you brainstorm ideas and see examples. All material you submit must be your own.”
“Use of AI such as ChatGPT is fully permitted, but you must cite the tool and be able to explain any work that you submit.”
5) Schools and departments should discuss these issues as the tools and their use evolve. OTEAR is available
for consultation and discussion at any time via email.
ChatGPT (Chat Generative Pre-trained Transformer) is a software application that is designed to have online chat conversations. It is a natural language processing tool that models human-like conversations. As a large language model it is described as powerful, and it remembers previous prompts so engaging with it emulates the give and take of human interactions. ChatGPT can answer questions and assist in writing tasks (see Vanderbilt email) such as emails, essays, reports, poetry, plays, and computer code. There are many potential applications for this artificially intelligent tool.
ChatGPT is prone to “hallucinations” where it “...writes plausible-sounding but incorrect or nonsensical answers.” When asked to cite sources, ChatGPT frequently fabricates this information and invents references that may appear legitimate, but are entirely fictitious. While the tool can be used to mimic many different writing styles, a great deal of initial guidance and proofreading is required before we should have confidence in the output.
There are many and various resources available to students to support their study, from calculators to a collection of AI tools. Faculty are encouraged to consider what their policies for dealing with these extraneous tools and provide clear instructions about which tools and resources can be used, how to cite and acknowledge them, discuss them in class and point out the policies on the syllabus, and remind students regularly of Rutgers Academic Integrity policy.
Develop clear policies for each course. For example:
“Use of AI such as ChatGPT is not permitted in any stages of the writing process on any assignment.”
“Use of AI such as ChatGPT is only permitted to help you brainstorm ideas and see examples. All material you submit must be your own.”
“Use of AI such as ChatGPT is fully permitted, but you must cite the tool and be able to explain any work that you submit.”
There are multiple different tools being developed to detect the use of AI generated content. OpenAI is also working on creating a “digital watermark” for all text created with ChatGPT. Here are some current detectors available:
However, it is unknown whether these tools can be effectively used for reviewing student work and enforcing academic integrity standards. As AI tools are constantly evolving, detection software may be less effective in the future. Here are some articles that further discuss these detectors, other tools, and their implications:
One strategy that can be used to discourage use of ChatGPT and promote learning is to require students to show stages of their work, e.g. outlines, rough drafts, selected references, contributions by different group members. Major assignments can be designed to be submitted in a sequence of steps, which is known as “scaffolding.”
Another strategy is that students can be required to add personal comments such as reflections on the value of the assignment in their education, what was challenging, surprising, or what they would do differently once the project was completed.
Instructors can also ask that students cite class readings or discussion, or require that they use recent scholarship which may not be available in AI databases. Or, students may be asked to collect data on their own and use it as part of their analysis.
ChatGPT can be used in a variety of ways in college courses: practice language skills when learning another language; use to organize ideas in a first draft; finding and organizing information in research papers; generate opposing arguments for discussions; create questions for content knowledge self-assessments; use as personalized tutor for improving learning. Many colleges are incorporating AI applications to communicate with students.
Take a look at how ChatGPT answered the question: How can ChatGPT be used in college courses?
ChatGPT can be used in a variety of ways in college classrooms, some examples include:
1. Language Learning: ChatGPT can be used to help students practice their language skills through conversational
interactions. The model can generate responses in the target language and can also be used to evaluate student's responses for grammar and fluency.
2. Essay Writing Assistance: ChatGPT can be used to help students generate ideas for writing assignments and
to provide feedback on their writing.
3. Research Assistance: ChatGPT can be used to assist students in finding and organizing information for research projects.
It can be used to generate summaries of articles, provide relevant citations, and can even be used to help generate research questions.
4. Group Discussion and Debate: ChatGPT can be used to generate discussion points and counterpoints for class debates and discussions.
5. Assessment: ChatGPT can be used to create multiple choice question for assessments and quizzes, which can help in reducing human effort and increase efficiency.
6. Personalized Tutoring: ChatGPT can be used to provide personalized tutoring to students, by answering questions and providing feedback on their work.
These are just a few examples of how ChatGPT can be used in college classrooms. As a powerful language model, ChatGPT has many potential
applications, and the possibilities are limited only by the creativity and imagination of the users.
The great success of ChatGPT has created a surge for other AI tools. Here is a list of some other AI applications available:
Now the Humanities Can Disrupt “AI” — Lauren M. E. Goodlad, Distinguised Professor of English and Comparative Literature, chair of Critical AI at Rutgers
AI Tools for Teaching and Learning – A Petting Zoo (Virtual Interactive Workshop)
Monday, June 12, 2023, 11:00am-12:30pm Host:AI Round Table
We've been discussing AI concerns and issues. Let's also discuss promising strategies for the productive, creative, and ethical use of AI in our classrooms and with our students. Join us and special guest Associate Professor Joyce Valenza for an AI petting zoo, where we will explore and test-drive applications for productivity, research, and instructional design and consider the framing of new norms for integration the coming semester (You can learn more about Joyce’s work here).
If you wish to attend the workshop or to receive the video recording, please register here.
Recommended Recordings
Academic Integrity and AI
The Chronicle of Higher Education, Tuesday, May 16, 2023
Host: Ian Wilhelm
Recording available here.
Since emerging in November 2022, ChatGPT has started numerous conversations regarding how Artificial Intelligence will change the future of work. When it comes to the classroom, ChatGPT has the propensity to churn out acceptable college-level essays and other assignment-related content. With this new innovation, what should the response be from instructors to AI-proof their classrooms? Join The Chronicle as they convene a panel of instructional experts to discuss what ChatGPT means for academic integrity and explore the other opportunities machine learning and artificial intelligence tools can create for plagiarism. This hour-long discussion will provide you with expert insights to navigate this unforeseen landscape in academic integrity, and stop instances of cheating before they become problematic in your classroom.
Unlocking the Power of AI: How Tools Like ChatGPT Can Make Teaching Easier and More Effective
Harvard Business Publishing Education, Tuesday, May 2, 2023
Speakers: Dr. Ethan Mollick & Dr. Lilach Mollick
Recording available here.
Artificial intelligence (AI)—including generative AI tools like ChatGPT—can be a force multiplier for instructors. Join Wharton Associate Professor Ethan Mollick and Wharton Interactive’s Director of Pedagogy Dr. Lilach Mollick as they showcase how AI can revolutionize the way you teach. In this webinar, you will: witness a live demonstration of working with Large Language Models (LLMs), learn how to use AI to generate useful teaching materials with ease, discover the advantages and disadvantages of various AI techniques, understand how AI can help identify knowledge gaps, uncover misconceptions, and support student learning.
How AI & ChatGPT are Changing Education – A conversation with Inside Higher Ed’s Susan D’Agostino
Susan D'Agostino, PhD is an educator turned journalist who covers technology issues for Inside Higher Ed. This year, she has been closely following developments in generative AI and education around the country and at international conferences, and is willing to share her insights with us. Join us for a conversation, facilitated by OTEAR, where we ask Dr. D'Agostino about issues of concern related to technology and the university.
ChatGPT Possibilities and Limitations in a Pedagogical Context (Interactive Live Demonstration)
ChatGPT is a powerful generative AI tool that scans the texts of the Internet and returns answers to questions that are detailed and articulate, although often inaccurate or superficial. Since its release in late November 2022, it has been the subject of much discussion, especially amongst educators fretting about the integrity of their take-home assignments. In this interactive demonstration, we will feed ChatGPT a variety of prompts to find out what it does well and what it does poorly. This session is meant to assist faculty in designing written prompts, reviewing student work, and understanding this new landscape.
Unpacking the Ethics of AI in Educational Development
Centering Center: POD Network Podcast, March 3, 2023
Podcast is available here.
In our last episode, we learned about ChatGPT with Cynthia Alby. Today we are diving deeper into the ethics of Al and how we as educational developers might best support the conversations happening across our campuses. To guide us in this conversation, we are speaking with Dr. Kevin Yee, Director of the Faculty Center at the University of Central Florida. Kevin has held numerous faculty and leadership positions at a range of institutions through his career including USF, Duke, lowa, Pomona College and UC Irvine. His research interests within pedagogy are wide, and have included student motivation, study skills, and various emerging technologies for teaching. He is currently co-editing a book of case studies on the intersection of VR and ethics in the college classroom.
AI and the University: Big Questions about Integrity and our Academic Future
Office of Teaching Evaluation and Assessment Research, March 1, 2023
Hosts: Chris Drue and Monica Devanas Panelists:
Michael Zwick, Senior Vice President for Research
Barbara Bender, Senior Associate Dean for Academic Support and Graduate Student Services
Joseph Deodato, Discovery Services Librarian, Rutgers University Libraries
Sharon Stoerger, Assistant Dean for Programs and Assessment, School of Communication and Information
New and ever improving AI technologies have changed the landscape of education. In this panel discussion we hear from representatives of key university units about the challenges that these tools pose for traditional educational practices and the opportunities they present for students and faculty alike. We will consider the ethical implications of AI in instruction, how faculty might responsibly use AI in their work, AI and the publishing industry, and how our university and other institutions should respond.
AI in the Classroom: Opportunities and Challenges
Lumen Circles, February 22, 2023
Derek Synder and Michael Babcock
Recording available via Zoom, Dereck Snyder's slides from session, and Michael Babcock's slides from session.
The release of ChatGPT late last year was immediately hailed as a game-changer in the application of artificial intelligence to writing tasks. Some academics have gone so far as to declare, somewhat apocalyptically, that “the college essay is dead.” This webinar takes a “technology-positive” approach to the challenges posed by A.I. to both humanities and STEM disciplines. The focus will be on specific, practical ways to use active learning techniques and process-based assessment to counteract the potential for cheating, while demonstrating how A.I. can function as a supportive technology in both the classroom and workplace.
AI FUTURES: An Interdisciplinary Conversation on Large Language Models and the Future of Human Writing
Rutgers Critical AI, February 20, 2023 Moderator/Introducer:Carter Mathes (English, Rutgers)
Video Recording available here on Youtube.
Virtual Roundtable with Kyle Booten (English, University of Connecticut), Sam Bowman (Data Science/Linguistics/CS, NYU – Visiting Researcher, Anthropic AI), Liz Losh (Rhetoric/American Studies, William & Mary), and Nasrin Mostafazadeh (AI Researcher, Co-Founder Verneek). Working from interdisciplinary standpoints in academia and industry, our roundtablers take up this topic in the wake of OpenAI’s recent release of ChatGPT, and its banning by NYC schools. Join our experts as they discuss the underlying technology of LLMs, their potential uses and abuses, and their likely impact on the future of human writing.
AI and the Academy: The End of the Essay?
Acadia Univresity, McGill University, January 31, 2023
Daniel Lametti - Associate Professor of Cognitive Psychology
Recording available on Youtube, Dr. Lametti's slides from session.
In this talk, I’ll review the strengths and weaknesses of large language models like ChatGPT to demonstrate that they do a poor job of completing most university assignments without a knowledgeable human hand to guide them. I’ll argue that these models are not a threat to higher education but a useful pedagogical tool that can help students learn how to write better papers, and facilitate more meaningful real-world interactions between students and professors. Finally, I’ll explore how universities might use AI to improve the experience of students in and outside of the classroom. In the future, language will be the mechanism by which humans interact with computers. Universities should prepare students for this change.
Developing Classroom Policies and Practices to Respond to ChatGPT and AI Assistance
Office of Teaching Evaluation and Assessment Research, January 10, 2023 Hosts: Chris Drue and Monica Devanas
Video recording available on the OTEAR Learning and Teaching Canvas site.
How should we respond to the emergence of AI platforms that allow students with an internet connection to automatically craft competent responses to a variety of prompts typically used in essays, homework assignments, reflection prompts, or exams? Should we avoid any mention of these tools in our classes, teach students how to use them responsibly, or ban their use outright? Is take-home work even valuable anymore? What implications does this have for grading and plagiarism? You have questions and we do too. In this discussion we explored what this new technology does and provided a space to voice concerns and solutions.
You are using a web browser that does not support
"CSS". Please ignore the part of this page below this text. You may be
using an older version of web browsing software that is likely to have
severe security and identity-theft issues, as well as problems displaying
certain web pages. You should download a newer web browser from Microsoft
or Mozilla.