ChatGPT: A Threat to Higher Education?

Colleges and universities are at an inflection point, grappling with declining enrollments, soaring costs, growing popularity of educational alternatives, and a dwindling perceived value of a degree (all of which I examine in my latest book). The latest threat, if you believe the hype, is ChatGPT, an AI-powered chatbot from Silicon Valley startup OpenAI. Released on November 30, the large language model (LLM) software has been predicted to do away with higher education (as well as journalism, code writing, and even Google, according to Gmail developer Paul Buchheit, who tweeted on December 1 that the search engine “may be only a year or two away from total disruption”). As someone deeply familiar with the crises facing my institution and others, I can say with certainty that ChatGPT is more hype than harm—and it might even offer some help.

A recent NPR story described ChatGPT as?“thrill[ing] the Internet with its superhuman abilities to solve math problems, churn out college essays and write research papers.” “Education may never be the same,” bemoaned RealClearEducation. Current and former teachers see danger signs, including Beverly Pell, an advisor on technology for children and a former teacher, who opined, “This could be a danger in education, and it’s not good for kids.” Writing coach and founder of Crush the College Essay Peter Laffin warns that ChatGPT and other AI chatbots “will lead to a crisis in learning, forcing educators to rethink schooling entirely.”

What these doomsayers fail to acknowledge is that programs and services for solving math problems and writing college essays and research papers already exist—and have for some time. The Photomath app, for example, has been around since 2014, with over 300M downloads, allowing students to take a picture of their math problems and get the answers. On the humanities side, papers have been for sale for years. Educators and admissions officers are not only well aware of these “services,” but have learned how to spot and work around them.

As I wrote in an Inside Higher Ed OpEd, “The key to retaining the value of a degree from your own institution is ensuring your graduates have the skills to change with any market. This means that we must tweak and adapt our curriculum at least every single year.” Those adaptations include a rigorous and ongoing survey of new technologies and services that could aid students in plagiarism and other forms of cheating, and tweaking homework and other assignments to reduce the effectiveness of those technologies. Many teachers—and anti-cheating software developers—are already doing this, and there is no reason to believe they will be stymied by ChatGPT.

OpenAI itself is working on solutions, according to?TechCrunch. University of Texas, Austin computer science professor Scott Aaronson, a guest researcher at OpenAI, said the company is “studying hiding cryptographic signals, called watermarks, in ChatGPT results, so that they’ll be more easily identifiable by [anti-cheating software] companies like Turnitin.”

The Possibility for Good

Once we move past the fear-mongering, it’s not difficult to imagine how LLM software can be used to enhance higher education. Ian Bogost, writing in The Atlantic, says “LLMs are surely not going to replace college or magazines or middle managers. But they do offer those and other domains a new instrument—that’s really the right word for it—with which to play with an unfathomable quantity of textual material.”

What can we do with that new instrument? Imagine asking students in a composition course to edit and fact-check a Chatbot-generated essay. Bogost says because the software “is simply trained to generate words based on a given input...it does not have the ability to truly comprehend the meaning behind those words. This means that any responses it generates are likely to be shallow and lacking in depth and insight.” Students could be asked to critique those weaknesses, add insights of their own, and spot the inaccuracies that are all too common with ChatGPT.

Data scientist Teresa Kubacka decided to run her own experiment by asking ChatGPT about a “cycloidal inverted electromagnon,” which she made up. She tweeted the results, which not only sounded plausible, but were supported by citations—which turned out to be bogus. Kubacka says the experiment left her with “an intense feeling of uncanniness: I just experienced a parallel universe of plausibly sounding, non-existing phenomena, confidently supported by citations to non-existing research. Last time I felt this way when I attended a creationist lecture.”

Sam Altman, co-founder of OpenAI, has acknowledged the problem. “ChatGPT is incredibly limited, but good enough at some things to create a misleading impression of greatness,” Altman tweeted. “It's a mistake to be relying on it for anything important right now. it’s a preview of progress; we have lots of work to do on robustness and truthfulness.” Indeed, users are warned that ChatGPT “may occasionally generate incorrect or misleading information.” That comes as no surprise to Stack Overflow, a coding resource forum that recently banned?ChatGPT because it would not stop giving the wrong answers.?

The New York Times’ Peter Coy says getting the most out of ChatGPT means acknowledging the limitations while “accentuating the positive.” He describes a YouTube video by web developer and product reviewer Jason Fleagle, who asked it to create a complicated web app. “The code wasn’t perfect,” Coy notes, “but Fleagle said, ‘As you can see, I just saved myself, like, a lot of time.’ There are dozens of such examples. ChatGPT can even?rewrite?software into a different programming language.”

Ethan Mollick, a professor at the Wharton School at the University of Pennsylvania's of Business, sees great potential as a “learning companion.” He told NPR he has used ChatGPT to help develop a syllabus, lecture, assignment, and grading rubric for MBA students. “It's this multiplier of ability, that I think we are not quite getting our heads around, that is absolutely stunning,” he said.

A multiplier of ability sounds like exactly what we want from new technologies. It may come with the potential to generate fake information and aid and abet cheating, but that’s nothing new, and nothing we can’t work around. Yes, as I’ve written before, higher education must change or die. But responding to the so-called threat that Chatbot software represents is not a concern.

* Originally published in?Forbes.