Generative Artificial Intelligence (AI) is consuming a growing share of our collective consciousness. Like many educators, we’ve spent months processing the media coverage and pondering AI’s potential impact on the future: Is AI “an absolute systems-level threat to education” as some argue? Will it spell the end of countless jobs or create new horizons for teaching, research, and meaningful work? Will AI save or enslave us?
It’s equal parts exciting and exhausting. But as educators, it’s important to remember we aren’t the only ones struggling to find our footing.
Survey data from Best Colleges reveals that as of March this year, 22 percent of students admitted using ChatGPT to complete an assignment or exam, with 57 percent indicating they intend to use AI tools for future coursework. That’s enough to give most instructors pause and has set off a flurry of activity in reimagining how we assess learning in order to reduce the risk of student cheating. But more far-reaching for the emotional beings who fill our classrooms are the concerns expressed about the future: Twenty-seven percent of students worry about AI’s influence on their education, with four in 10 saying AI defeats the purpose of education altogether. Thirty-one percent are fearful about their career prospects and almost half are anxious about AI’s impact on society. A survey by ZipRecruiter adds even more gravity to these concerns: Seventy-two percent of Gen Zers say they are worried about losing their jobs to AI while The Economist found that Google searches for “is my job safe” have doubled in recent months.
The troubling part is that, according to Best Colleges, the majority of students report their instructors have not openly discussed the use, let alone potential impact of AI on education, careers, or the very disciplines students are investing their time and treasure in pursuing.
Furthering a culture of care in the classroom
There’s plenty of research demonstrating how stress interferes with learning, especially in memory formation and retrieval. Meaningful, productive dialogue can help. By exploring the concerns AI is eliciting in our students, we can also create the space to address the anxiety many of us feel about the issue of cheating.
Earlier this spring, Demian Hommel, co-author of this article, devoted a class to discussing how the risks posed by AI might affect students’ personal and professional aspirations and what they could do to enhance their own resilience in the face of change. The responses reveal a mixture of apprehension and optimism. Most noteworthy, they show a hunger for dialogue:
- “AI can pretty much do anything right now and it’s really threatening many jobs.”
- “Advances in technology are inevitable. People have resisted new and drastic change throughout history, yet every time it happens we seem to adjust as a society.”
- “I’m taking four courses this term and this is the only time that the impact of AI has been brought up, even though my other courses are computer science and engineering where this technology is going to be most disruptive. It’s true that we can’t predict where this is all going, but I feel better just having discussed it.”
Whether or not we’ve wrapped our minds around AI, or determined how it will affect the work we ask students to shoulder, at the very least, we need to have a conversation.
Setting the table for an effective discussion
Since most of us don’t have all the answers, this moment offers a unique opportunity to model vulnerability, intellectual curiosity, and the willingness to learn alongside our students. As Marie Curie famously said, “Now is the time to understand more, so that we may fear less.” But to avoid a reactionary discussion, it’s important to set the stage. Thinking in terms of before, during and after class is a useful framework for this or any other substantive discussion we may wish to have with students.
To prepare for the discussion, assign readings that provide different perspectives on AI along with comprehension questions for students to complete before class. Danny Liu and Adam Bridgeman’s recap from a recent student forum held at the University of Sydney offers a host of areas for exploration. The Best Colleges Survey might also serve as a foundation by exposing students to how others are responding to AI—from metrics on student usage, to perspectives on academic integrity, education, society and the world of work. Even better, have students complete a survey of your own creation delving into these themes along with questions specific to your discipline and related career pathways.
Preparing to engage
Getting students to open up can be challenging. Asking a series of questions at the start for students to reflect on individually or with a partner (“think-pair-share”) will increase the likelihood of participation. For large classrooms, technology is a powerful enabler, especially for those who may be less inclined to speak up. Today’s student engagement platforms offer a range of options to gather responses through polling, word answer and other question types and to display them in a visually engaging way.
You might ask what skills or knowledge students believe are critical in the age of AI? Or how they might feel if their instructor used ChatGPT to provide feedback on their assignments? Single word responses can be turned into a word cloud to spur discussion. Click-on-target questions allow students to use their laptop or mobile device to click on a portion of an image, a graph or even a paragraph, which then converts individual responses into a heatmap. For example, Hommel used this approach to show an infographic of the world’s largest economic sectors and asked students to select the industry they believe will be most affected by AI. The results—that nearly every part of the economy is likely to be affected—were surprising to some students who had not considered the role information technology plays in each sector.
If you have a tool that allows you to host discussions, create a thread and keep it open so students can ask questions and provide comments. This allows students to generate conversations with their peers and to surface concerns and opinions from individuals and the collective. Keeping tabs can be challenging, so consider enlisting the help of a teaching assistant or student to report in periodically.
If you create your own version of the Best Colleges survey, reveal the results at key moments and then ask for opinions and counterpoints. You might also compare and contrast the results against those from your own students and use this to explore the similarities and differences. Having students react to the opinions and perceptions of their peers is a surefire way to generate meaningful engagement.
Power in reflection
Although sharing our own perspectives may help, the most meaningful driver of academic and personal growth are the insights students generate for themselves through reflection.
At the end of the discussion, have students write an exit ticket or a minute paper to capture their most important takeaways, any shifts they’ve experienced in their attitudes towards AI, and what they might do moving forward. This offers instructors the benefit of understanding where students are at, if the experience was beneficial, and what concerns might be left unaddressed. If an end-of-class assignment seems too rushed, consider assigning a series of reflection questions for extra credit. You might ask them to conduct research on an aspect of AI they’re interested in exploring. What opportunities does AI afford to revolutionize traditional approaches or methods in your field? How might students use generative AI to make aspects of their own lives more efficient? Or as Liu and Bridgeman asked, if you use AI to complete assignments, how will future employers have confidence in your own knowledge and capabilities?
One critical benefit of the before, during and after approach is signaling the value of the collective learning that happens in the classroom. Asking students to prepare ahead of time ensures they are less likely to be passive participants and will improve the quality of the discussion. Leveraging technology for the during portion helps each student see their response in the context of the larger picture. Following up with a reflection activity afterwards makes it more likely the information and experience will “stick.”
Putting students in the driver’s seat
Anxiety is often a product of uncertainty. Encouraging students to reflect and think critically about their thoughts and emotions and the actions they might take can help them separate perceived from actual threats. Given some of the apocalyptic predictions about AI’s impact on society, providing a venue for thoughtful dialogue allows us to model the value of processing major issues as a collective. It also gives students the opportunity to inform their own opinions and plans for the future, which may engender a greater sense of control over events that impact their lives.
The same advice applies to those of us charged with leading students through their academic journey. We’re all grappling with uncertainty and the sheer volume of recommendations and prognostications surrounding AI is overwhelming. In the face of a change as significant as this, we would be the first to recommend investigating how AI will affect course delivery and familiarizing ourselves with the potential of these tools to transform learning. But we should also be having the same discussion with our peers and within our departments. Sharing our own encounters with students about the opportunities and concerns raised by AI is a good place to start.
(Generative AI was not used to write this article)
Dr. Demian Hommel, PhD, teaches introductory and upper-division human geography courses in the College of Earth, Ocean, and atmospheric sciences at Oregon State University. He is also a fellow for the institution’s Center for Teaching and Learning, working to push the mission of excellence in teaching and learning across his campus and beyond.
Dr. Bradley Cohen, PhD, is the Chief Academic Officer at Top Hat where he provides leadership and advocacy for personalized, inclusive and equitable teaching practices within the higher education community. Prior to joining Top Hat, Cohen served as the Chief Strategy and Innovation Officer at Ohio University and as the head of the Center for Educational Innovation and Associate CIO for Academic Technology at the University of Minnesota.
References
Admin (2023). This Time is Different. Part 1. Elearnspace.
Welding, L (2023). Half of College Students Say Using AI on Schoolwork Is Cheating or Plagiarism. Best Colleges.
Lee, J (2023). Effective assessment practices for a ChatGPT-enabled world. Times Higher Education.
Trust, T (2023). Essential Considerations for Addressing the Possibility of AI-Driven Cheating, Part 1. Faculty Focus.
ZipRecuriter (2023). The ZipRecruiter Job Seeker Confidence Survey. ZipRecruiter.com.
Economist (2023). AI is not yet killing jobs. Economist.com.
Hobson, N (2018), Why Your Brain on Stress Fails to Learn Properly. Psychology Today. Psychologytoday.com.
The Editors (2023). How Will Artificial Intelligence Change Higher Ed? The Chronicle of Higher Education. Chronicle.com.
Liu, D., & Bridgeman, A. (2023). Students answer your questions about generative AI – part 1: Assessments and their future. University of Sydney. Educational-innovation.sydney.edu.au.
Top Hat. Think-Pair-Share. Glossary of Higher Ed. Tophat.com.
Top Hat. Exit Ticket. Glossary of Higher Ed. Tophat.com.
Top Hat. Minute Paper. Glossary of Higher Ed. Tophat.com.
