This study examines the integration of ChatGPT into an advanced Korean course in a U.S university context, with a focus on its instructional design and classroom implementation. Over a semester, eighteen students used ChatGPT as a supplementary tool within process-oriented writing tasks. Survey data were used to examine learners’ responses to this instructional use of ChatGPT. The results indicate that students perceived ChatGPT as particularly helpful for grammatical corrections and receiving immediate feedback. Learners who recognized its benefits reported a positive impact on the usefulness of feedback, their writing development, and interest in future AI use. However, concerns about over-reliance, ethical issues, and occasional inaccuracies were also noted. By situating ChatGPT within a specific instructional context, this study highlights both the perceived benefits and limitations of AI-assisted writing in Korean language education. This study also offers practical guidance for educators, including structured stages for AI application in writing, usage guidelines, assessment criteria, and sample prompts. Finally, it emphasizes a balanced approach in which AI complements, not replaces, teacher feedback to maintain quality through professional judgment.

This qualitative study explores how fifth-grade elementary school students experience AI-based feedback in English writing. Two students participated in writing activities and received feedback from ChatGPT. Data were collected from observation logs, drafts, reflection journals, and interviews over several weeks. The analysis focused on participants’ perceptions, cognitive/meta-cognitive and affective responses, and the impact of feedback on revision. Results indicate AI-based feedback is helpful, clear, and supports self-directed learning, increasing motivation and confidence. Effective classroom use requires instructional sequences integrating AI feedback with teacher facilitation, including drafting, feedback reception, revision, and reflection stages. Teacher training should prepare educators to manage AI tools effectively and foster students’ self-regulation. Teacher-led research and collaboration can enhance AI feedback integration, ensuring meaningful learning experiences for young EFL learners. These findings provide foundational insights for developing personalized feedback systems and instructional models combining technological innovation and pedagogy, supporting young learners’ growth as confident and autonomous writers.

This study investigates how artificial intelligence (AI)-generated and instructor-provided feedback interact within process-based, collaborative EFL writing to influence learners’ affective engagement, motivation, self-efficacy, and attitude toward writing. Semistructured group interviews with nine pre-service English teachers were analyzed using NVivo 14. Among the three categories, attitude, encompassing both positive and negative factors, was the most frequently observed, followed by self-efficacy and motivation. Positive attitude was linked to the convenience, immediacy, and personalization of AI feedback, and to the collaborative course design that supported iterative revision and peer learning. Negative attitude reflected concerns about AI overreliance, inaccuracies, and limited contextual appropriateness, particularly in ESPfocused writing. Regarding self-efficacy, instructor feedback provided deep, contextaware guidance, whereas AI feedback offered rapid, mechanical accuracy and motivational support. Confidence improved through safer drafting, metacognitive insight, and gradual mastery of AI utilization skills. Motivation was reinforced by the process-oriented structure, targeted instructor feedback, collaborative activities, and peer support that fostered engagement and improvement.

This study investigates how AI feedback learning activities in English writing classes affect university students’ English achievement and writing motivation. It examines both cognitive and affective outcomes in EFL contexts using generative AI tools such as ChatGPT. A total of 244 first-year university students completed four writing tasks incorporating AI feedback during the revision process. Pre- and post-tests were conducted along with the Oxford Placement Test (Syndicate, 2001) and a modified version of Payne’s (2012) Academic Writing Motivation Questionnaire. Quantitative analyses revealed significant improvement in English achievement and three motivational subcomponents—enjoyment, recognition, and effort—while instrumentality showed no significant change. Qualitative analysis of open-ended questionnaire responses indicated that students valued AI feedback for error detection, coherence, and self-directed learning, yet recognized limitations such as unnatural expressions, misinterpretations, and overreliance on AI. The findings highlight the importance of AI literacy and propose a hybrid feedback model integrating AI and teacher guidance to enhance learner autonomy and writing competence.

본 연구는 Deepseek와 ChatGPT가 중국 대학 한국어 학습자들의 쓰 기 텍스트에 대해 제공하는 피드백 양상을 살피고, AI 피드백의 활용 방 안을 모색하는 데 목적이 있다. 이를 위해 한국어 학습자 30명을 대상으 로 쓰기 텍스트를 수집하고, 두 도구가 텍스트에 대해 제공한 피드백을 분석하였다. 분석 결과, 두 도구 모두 평가 기준에 기반하여 피드백을 제 공하고, 학습자가 수용하기 쉽도록 문제점과 개선 방향을 명확하게 제시 하였다. 대부분의 피드백은 정확도가 높은 편이었으나, 평가 범주와 오류 유형이 부합하지 않는 경우도 일부 나타나고 있었고, 특히 Deepseek의 경우 쓰기 주제를 잘못 파악하여 내용적 측면에서 부적절한 피드백을 제 공한 사례가 많았다. 두 도구 모두 쓰기 텍스트의 내용에 따라 피드백을 제공하여 우선순위는 두드러지지 않았고, Deepseek가 ChatGPT에 비해 지지적인 어조를 더 많이 사용하는 것으로 나타났다. 본 연구는 생성형 AI를 활용한 한국어 교육 및 연구를 위해 기초적인 자료로 사용할 수 있 을 것이다.

이 논문은 S대학교 ‘사고와 글쓰기’ 교과에서 AI 자동 피드백 프로그 램 키위(KEEwi)를 활용한 수업 사례를 분석하고 그것이 지닌 효과와 한 계에 관해 탐색할 것을 목적으로 한다. 먼저 2장에서는 AI 자동 피드백 프로그램의 동향과 원리를 살펴보고 3장에서는 키위 프로그램을 활용한 수업 운영과 구체적 사례에 관해 설명하였다. 4장에서는 설문조사 분석 을 통한 학습자들의 프로그램 만족도와 교육 효과 및 한계에 관해 논의 하였다. 그 결과 이 프로그램은 즉각적이고 반복적인 피드백을 통해 학 습자의 학습 능률을 향상시킬 뿐 아니라 수업에 대한 흥미를 유발하여 글쓰기 역량을 강화하는 데 긍정적인 영향을 끼치는 것으로 드러났다. 또한 교수자에게는 글쓰기 교육의 보조 도구이자 협력자로 기능하며 학 습자 중심의 피드백 환경을 조성하는 데 기여하는 것으로 파악된다. 그 러나 키위 엔진은 완전한 대체 도구가 아니며 자체적인 오류와 문제를 가지고 있어 절대적인 평가 지표로 활용하는 것에는 한계가 있음이 확인 되었다. 이러한 연구는 대학 글쓰기 교양수업에서 AI 자동 피드백 프로 그램이 학습자와 교수자에게 끼치는 영향과 그 한계를 파악하는 데 도움 이 되리라 생각한다.