Artificial intelligence in education can offer equitable learning opportunities for less-advantaged students, but it may not alleviate the intense competition among elite students striving to gain admission to prestigious colleges, according to a US education technology pioneer.

“AI appears to be most useful to people who aren’t really good. If you’re a great writer, using AI to write your article (wouldn’t be helpful). But AI can help the bottom of the class and, arguably, you can make a case that will make education more fair,” John Katzman, co-founder of The Princeton Review, said in an interview with The Korea Herald on Monday.

South Korea’s shift to AI-driven digital textbooks in 2025, in particular, could be a tremendous force in making learning more engaging, he said, highlighting that at the core of AI education is allowing students to study independently on their own schedule.

“(AI education) should be different for every student because it’s not about what happens in the textbook. It’s not about how (teachers) should get (students) ready for a class on Thursday,” he said.

Calling himself a proponent of AI education, Katzman said active learning through digitalized lessons presents students with a problem that it will help them solve -- and give them a reason to want to solve it.

“It’s not just saying ‘you go learn stuff’ to students in a lecture. Keeping schools the same as they were 50 years ago when the rest of society is changing just means it’s less relevant.”

Katzman, however, said parents’ and students’ reliance on private academies will remain course even if AI learning tools provide one-on-one lessons and tutor students intelligently.

“We still need teachers, and they can be more effective because they’re backed by AI tools. I don’t know why it would be any different in the tutoring world,” he said.

Katzman said The Princeton Review made tutors more efficient by giving them better teaching tools, and students also did not get less tutoring just because AI streamlined teaching.

“They did exactly the same amount and got more out of it. If it’s more efficient, then students will just learn more,” Katzman said, noting that AI digital textbooks will not stop students from flocking to the private sector.

To cool off the vibrant education culture, Katzman said universities in Korea should change their candidate selection method, as they currently place heavy emphasis on stellar academic grades, causing grade inflation. Applications with all A’s should not rise to the top of a pile in an admissions office, which would in turn sink the chance of getting into a prestigious college for others, he added.

Katzman said students are also missing out big on learning because of the “grade matters” policy.

“Think of college admissions as a casting coach for a play. I don’t need the five best actors in Seoul. I need somebody who’s going to play the specific role,” he said. “It’s not about the extra hours of test practice.”

Since colleges are not looking for academic machines, they should look for students to compose a diverse student body, which Katzman believes would shift students away from working toward making their application stand out among the admissions pool with good grades to focusing more on engaging in other activities.

“Why work hard? We want you to have a life. What gives you joy? Where do you want to spend your time? If you’ve got extra energy, spend it elsewhere, not memorizing like crazy. But students don’t see that. All they see is ‘I got in’ or ‘I didn’t get in,’” Katzman said.

“AI can’t solve (the cutthroat competition of education) because the incentives are wrong at the elite level. If you just see AI impacting the textbook, you’re missing a big story.”