Dr. Bouchey is Associate Professor and Vice Provost of Digital Strategy and Operations at National Louis University where she is responsible for standards of quality and service for online programming across the institution, alternative graduation pathways, learning experience design, and academic innovation, overall. Dr. Bouchey holds a B.A. in Psychology from the University at Albany, an M.B.A. in Entrepreneurship from Rensselaer Polytechnic Institute, and a Doctorate in Education from Northeastern University. Her research interests include the nature and future of organizational structures of online units in institutions of higher education, as well as inventive and high-impact pedagogical practice in online teaching, inclusive of artificial intelligence and machine learning. Dr. Bouchey writes and is widely quoted in the academic and popular press; her articles and curriculum vitae can be accessed here: www.drbouchey.com.
The world of higher education is rapidly evolving, and technology has played an important role in that process. Technology has enabled educators to teach, learn, and collaborate with students in new ways, including opening the walls of the classroom to digital and online education. One recent, exciting innovation at the intersection of technology and education is ChatGPT, which leverages artificial intelligence (AI) and has the potential to transform teaching and learning.
ChatGPT stands for “chat generative pre-trained transformer.” This AI system uses natural language processing to generate language in response to prompts or questions, much like humans having a dialog. ChatGPT, developed by a collaboration between Microsoft Research Asia and Peking University, uses AI to automate interactions between people and computers. Unlike its cousins Amazon’s Alexa or Apple’s Siri, this tool doesn’t ask for information—it simply learns from every dialogue in order to provide increasingly accurate responses. Although a who’s who of technology leaders is calling for a 6-month pause on AI training to give the world some time to catch up, there is little doubt that AI is here to stay.
Many of us in higher education have begun to imagine a world where some of the administrative tasks associated with teaching and learning could be accomplished (dare I say offloaded?) through the use of AI. One such task is wraparound student support. Intelligent agents added to data-informed student support models could identify at-risk student behaviors more quickly than a human with a data model could. This would free up advisors and other student support employees to intervene with individual students in thoughtful and informed ways. AI can help deliver data-driven insights into how courses (and students) are performing, and which topics should be focused on or revised. This data can be used by faculty and administrators to improve overall course quality and ensure students receive the best possible education.
Other critical student support systems could be scaffolded with AI. Automating statuses related to financial aid and generating scholarships that match details in financial aid applications would reduce the cost of an academic journey. Incoming students could upload a resume or answer questions about their previous work and life experiences so an AI tool could identify credit for prior learning in a more cost-effective way as well. Using AI to address the increasing pressure on the cost of higher education is a unique opportunity not only for students, but also for higher education to address rising costs as well.
AI could also help with time management for faculty. Uploading content or a course syllabus into a chatbot could reduce the need for oft-asked questions related to the course (e.g., when is the assignment due this week?). It could also help students find additional materials that augment the course or are at higher or lower taxonomies, as needed. In addition, commonly asked questions could be used to generate course improvement ideas and recommendations related to those improvements. Faculty and students could agree to use AI to generate more feedback on assignments and thereby increase the chances for student success without the correspondent increase in faculty workload.
Finally, AI could help with personalized learning. Online learning experiences could be made more interactive and engaging with the help of ChatGPT, allowing students to ask detailed questions about a topic for quick responses in real time. Personalized tutoring is achievable by having ChatGPT explain topics differently and perhaps through different modes as well (e.g., videos instead of text). If one explanation doesn’t make sense for a student, ChatGPT could provide another that may better explain what’s being taught to that individual student and to their learning preferences. As an added benefit, an AI tutor would not experience fatigue as human tutors would.
Opportunities notwithstanding, there are many outstanding questions and potential downsides to the use of AI in teaching and learning that we must also attend to. These are related to the call for the pause on AI training by the likes of Elon Musk and Steve Wozniak. For us in higher education, some of the chief questions relate to ownership and copyright. If a tool like ChatGPT formulates a poem, for example, who owns it—the machine or the human who published it? Further, because most of these AI tools are trained only on information openly available on the internet, the chance of spreading inaccurate information, even disinformation, is very high. Rampant bias in generated responses has been reported because these systems cannot access information located behind a login or paywall, such as peer-reviewed materials and textbooks written by vetted and more objective experts.
Moreover, the use of AI in writing calls into question the value of written assignments in higher education. While many educators have long lauded the value of multimodal learning environments that enable students to write, record, or draw to demonstrate their understanding of a key concept, the domination of writing in higher education cannot be understated. We don’t yet know how AI will disrupt our current assessment strategies, but it clearly is time to evaluate those notions.
AI and applications such as ChatGPT are truly revolutionary technology that can and will shift the teaching and learning landscape and higher education in general. Despite our incomplete understanding of the long-term benefits and consequences, AI will take center stage in the future of higher education. With all of this in mind, we must be listening, experimenting, and evaluating how AI will impact higher education today and into the future.