Overseas campuses embrace AI while Korean universities clamp down

gettyimagesbank

Cheating scandals involving artificial intelligence (AI) have continued to emerge at Korea’s top universities, prompting the government to move toward establishing the country’s first ethics framework for student use of the technology.

Beyond Korea, however, universities are embracing a different approach to AI, highlighting calls for a broader shift in how the country’s higher education system responds to the AI era.

“AI use has become impossible to fully restrict,” said Lim Woo-young, an economics professor at Hong Kong University of Science and Technology (HKUST).

He explained that the overall campus climate at HKUST is broadly supportive of AI use, with the university encouraging students to engage with the technology rather than banning it outright.

At HKUST, the university does not impose a single, institution-wide mandate on how faculty should handle AI in their courses.

Instead, individual professors are expected to establish clear standards on AI use and communicate them explicitly to students, including by outlining those guidelines in course syllabi.

Image of Hong Kong University of Science and Technology campus / Courtesy of Hong Kong University of Science and Technology

When generative AI first emerged, approaches varied widely. About two years on, however, most faculty members now actively encourage the use of AI.

“If AI is accepted as a tool for improving productivity, then the ability to use AI should also be reflected in how students are evaluated,” Lim said.

HKUST designs exams and assignments with AI use in mind, shifting assessments toward questions that demand deeper thinking, such as open-ended problems without a single correct answer or tasks in which students are evaluated on the clarity and logic of their reasoning rather than rote responses.

“The direction of social change is clearly toward greater use of AI,” he added. “Educators are now at a point where they, too, need to adapt.”

The University of Utah Asia Campus is taking a similar approach to AI use on campus.

“We don’t see AI as something to ban, nor something to rely on blindly,” said Gregory Hill, chief administrative officer of the University of Utah Asia Campus. “At the University of Utah, we view AI as a tool that can support learning.”

He said the university’s focus remains on the type of graduates it aims to develop, emphasizing core competencies such as critical thinking, communication, collaboration and adaptability — skills he said remain essential even as specific AI tools continue to evolve.

It follows university-wide guidelines and policies, which clarify how existing rules apply to the use of AI, particularly in areas such as data protection, research integrity and responsible use.

Rather than enforcing a single, uniform rule across all courses, the campus places emphasis on clarity at the course level. Like HKUST, faculty members are expected to clearly state how and whether AI tools may be used in their classes.

“We encourage faculty to rethink assessment approaches that make students’ reasoning and learning processes visible, rather than relying on a single technical fix,” Hill said.

Neither HKUST nor the University of Utah Asia Campus has experienced large-scale or organized AI-related cheating cases like those recently reported in Korea.

At HKUST, there have been isolated cases in which students used AI to assist with essay assignments or take-home exams. However, such one-off incidents did not escalate into broader controversies, as the university and individual faculty members adjusted their assessment methods in response by redesigning exams and assignments to reduce reliance on formats vulnerable to AI misuse.

“We now all recognize that it is extremely difficult to completely prevent students from using AI in essays or take-home exams, no matter how strict the checks are,” Lim said.

He mentioned that the recent string of cases in Korea reflected shortcomings in how universities and faculty responded at an early stage.

“In many cases, the best response was not in place,” he said. “Problems emerged because investigations were carried out only after incidents had already occurred, instead of through proactive changes to assessment and evaluation systems.”

A view of the University of Utah Asia Campus / Courtesy of University of Utah Asia Campus

Hill agreed that the challenges extend beyond Korea, but stressed that local context still matters, noting that in systems heavily reliant on high-stakes examinations, concerns over fairness and misconduct tend to be particularly acute.

“Our goal is not to train students for a single technology, but to prepare them for continuous change,” Hill said.

By strengthening critical thinking, ethical judgment, communication and adaptability and by encouraging responsible AI use in line with university policies, the university aims to equip students to engage thoughtfully with both today’s tools and tomorrow’s innovations.

Another factor behind the string of scandals at Korean universities has been the widespread use of online exams.

While many universities abroad have taken a relatively flexible stance on AI use, they have been far more cautious about online testing.

At HKUST, for example, even when classes are conducted online, objective assessments are typically held in person.

University officials note that there is no reliable way to monitor exams online, despite efforts such as mandatory camera use, angle and screen controls and exam-monitoring software. As oversight measures evolve, so do methods of cheating, making technical controls increasingly ineffective.

Korea’s top universities of Seoul National University, Yonsei University and Korea University have moved to tighten institution-level measures following a series of large-scale misconduct cases.

At the policy level, the Ministry of Education is also working with the Korean Council for University Education to develop AI ethics guidelines for university students.

Still, observers caution that efforts focused solely on preventing future scandals may fall short, as universities face growing pressure to rethink assessment models and adapt to the AI-driven transformation of higher education.