AI-powered Korean language learning boosts focus, reading comprehension, study finds

Daekyo and a research team led by KAIST professor Jeong Jae-seung have released findings verifying the effectiveness of AI-personalized Korean language learning. Students who received both reading passages and questions matched to their level showed the highest concentration and meaningful gains in reading comprehension.

Daekyo announced Tuesday that the Daekyo X KAIST Cognitive Enhancement Research Center, which it jointly operates with KAIST, published the findings on the educational impact of AI-tailored Korean language instruction.

The study examined how AI-personalized learning — providing reading passages and questions calibrated to each student's level — affects engagement and reading comprehension. The research center compared learning outcomes among 90 first-year middle school students based on whether they received personalized passages, personalized questions, both, or neither.

The analysis found that students given fully personalized conditions — both passages and questions matched to their level — showed the highest concentration. Reading comprehension also improved significantly. Comprehension was measured using KReaD (Korean Reading Index), a diagnostic assessment developed in-house by Daekyo.

The research center also analyzed long-term learning data from 1,582 middle school students in grades 7 through 9 who used Summit Suneung Korean Training, an AI-personalized Korean language learning program. The average study period for those students was 373 days. The center said overall academic achievement improved significantly regardless of students' initial reading level.

The findings come as the education industry moves to strengthen its competitiveness through AI-based personalized learning services, amid slowing growth in the traditional home-visit tutoring market driven by a declining school-age population. Daekyo posted consolidated sales of 157.7 billion won ($102 million) and an operating loss of 3.6 billion won in the first quarter of this year. Sales fell 4.8 percent from the same period a year earlier, and the company swung to an operating deficit.

Daekyo plans to use the research results to advance its AI-based personalized learning services and individual study design. The company said it will focus on quantifying each learner's reading level and improving engagement and achievement by providing passages and questions tailored to that level.

"This research is significant in that it scientifically validates, through KAIST's findings, that the personalized learning offered by Summit Suneung Korean Training — which applies an AI learning system — has a positive effect on learner engagement and achievement," a Daekyo official said.

"Personalized learning matched to a student's level is a key factor in boosting engagement and driving achievement," said Jeong, the KAIST professor. "Providing a learning environment suited to each learner's level is an effective educational strategy."