Earticle

This study creates and selects a senior image model suitable for senior hair-up style design by utilizing the strengths of prompt accuracy, design modification, and artistic consistency with interactive artificial intelligence, ChatGPT 5 (Open AI). Prompt-based hair-up styles were designed through a generative AI model. With the focus on natural and elegant hair wave styles, popular in 2026, this study intend to produce a total of 24 designs for senior hair-up styles by theme (Hair length-3 points, Hair Point-6 points, Hair Parting-9 points, and Hanbok & Dress-6 points). Quantitative and qualitative analysis was conducted by expert sensory testing on various senior hair-up style designs produced by image generating AI models, and objective sensory evaluation of the designs were conducted from the perspective of formability, dignity, suitability, and preference. As a result of the research on senior hair-up designs using ChatGPT 5 (Open AI), it was confirmed that AI can be used in various ways in the beauty field, including a creative design tool as well as a basis for practical work. This study intend to effectively apply AI to the senior hair-up style education environment and develop AI convergence class contents.

本研究核心内容分为三部分：第一，依托对话式人工智能ChatGPT-5（OpenAI）在提示词精准度、设计迭代、风格一致性上的优势，构建并筛选适配中老年盘发设计的人物图像模型；第二，基于筛选后的AI 模型，通过提示词驱动完成中老年盘发设计，以2026 年自然优雅卷发趋势为风格核心，围绕中老年盘发主题，按发长（3 款）、造型亮点（6 款）、刘海分缝（9 款）、韩服/ 礼服场景（6 款）分类，共完成24 款设计作品；第三，邀请行业专家对AI 生成的盘发设计开展感官评价，从造型性、端庄感、适配性、偏好度四个维度进行定量与定性评估。研究证实，运用ChatGPT-5 开展中老年盘发设计，不仅可辅助完成前期创意筹备，还能作为创作工具提升设计完成度，在美容领域具备多元应用场景。本研究提出的AI 辅助设计思路可应用于中老年盘发教学场景，用于开发美妆设计与AI 融合的课程内容。

This study explores the feasibility of integrating image-generating AI into a sustainable STEAM education, Science, Technology, Engineering, Arts, and Mathematics, with an arts-centered approach. Using Microsoft Copilot to conduct an AI image generation competition involving 206 participants from various demographic groups, we investigated creative engagement patterns and educational potential. Participants were tasked with producing images inspired by thematic keywords including ‘spring,’ ‘Korean traditional house,’ ‘robot,’ and ‘aerospace.’ The participants were divided into three groups for analysis: undergraduates within four years of admission, undergraduates who had been enrolled for four or more years, and non-undergraduate participants. The results revealed notable differences in creative tendencies across groups. Non-undergraduate participants demonstrated a more diverse use of colors and themes as opposed to undergraduate students enrolled for more than four years. The inclusion of certain lexical items, such as ‘space’, was significantly higher among specific groups. Text-to-image generation observed varied thematic preferences and engagement levels indicating possibilities for future resource design. These findings underscore the potential role of generative AI in fostering creativity, bridging art and technology, and promoting sustainability in education. This research highlights the transformative potential of AI tools in educational settings, offering insights into their potential application for personalized learning experiences. The implications extend to developing tailored educational programs that address diverse learner needs, ultimately contributing to the advancement of arts-integrated STEAM education.

In this study, we investigate the changes and realities of the content production process focusing on Image generation AI revolutions such as Stable Diffusion, Midjourney, and DELL-E, and examine the current status of related department operations at universities and Find out the status of the current curriculum. Through this, we suggest the need to produce AI-adaptive content talent through re-establishing the capabilities of content-related departments in art universities and quickly introducing curriculum. This is because it can be input into the efficient AI content development system currently being applied in industrial fields, and it is necessary to cultivate talent who can perform managerial and technical roles using various AI systems in the future. In conclusion, we will prepare cornerstone research to establish the university’s status as a source of talent that can lead the content industry beyond the AI content production era, and focus on convergence capabilities and experience with the goal of producing convergence talent to cultivate AI adaptive content talent, suggests the direction of curriculum application for value creation.

Focusing on the rapid progress of image generation AI, this study examines the changes in talent required according to changes in the production process of the content industry, and proposes an educational management plan for the subject and comparative department of the university’s animation major. First, through environmental analysis, the trend of the animation content industry is analyzed in three stages, and the necessity of producing AI-adapted content talent is derived by re-establishing the talent image of the university’s animation major and introducing it into rapid education. Next, we present a case designed by applying teaching methods to improve technology convergence capabilities and project-oriented capabilities by presenting subject and non-curricular cases operated in the animation department of the researcher’s university. Through this, we propose the necessity of education to cultivate animation content talent who can play technical and administrative roles by utilizing various AI systems in the future. The goal of this study is to establish a cornerstone study by presenting application cases and having the status of a university as a talent supplier that can lead the content industry beyond the era of AI content production that breaks the boundaries of genres between contents. In conclusion, it is intended to propose the application of education to create value through technology convergence capabilities and project-oriented capabilities to cultivate AI-adapted content talents.

최근 이미지 생성형 AI 기술의 발전은 디자인 분야에서 창의적 작업을 지원하는 도구로 주목받고 있다. 본 연구는 그래픽 디자인 분야에서 주요 이미지 생성형 AI 툴인 DALL·E, Midjourney, Stable Diffusion Online, Firefly의 활용 우위를 분석하는 것을 목적으로 한다. 이를 위해 아이덴티티 디자인, 포스터 디자인, 웹 디자인의 세 분야를 대상으로, 문헌 연구와 전문가 설문을 통해 각 AI 툴의 프로세스 단계별 적용성과 효과를 평가하였다. 연구 결과, 프로세스 2단계(아이디어 발상)에서는 DALL·E가, 3단계(디자인 시안 제작)와 4단계(디자인 선별 및 개선)에서는 Midjourney가 상대적으로 높은 평가를 받았다. 또한, Firefly는 초기 프롬프트 생성에서는 상대적으로 성능이 낮았으나, 프로세스가 진행될수록 평가가 점차 상승하여 레퍼런스 이미지를 기반으로 한 수정 작업에서 비교적 긍정적인 결과를 보였다. 본 연구는 디자이너들이 작업에 적합한 AI 툴을 선택하는 데 유용한 가이드를 제공하며, 디자인 프로세스에서 AI 기술을 효과적으로 활용할 수 있는 기초 자료로 활용될 수 있을 것으로 기대된다.

The rapid advancement of image-generating AI technology has garnered attention as a creative tool in the design field. This study aims to analyze the priority of major image-generating AI tools-DALL·E, Midjourney, Stable Diffusion Online, and Firefly-in graphic design. Focusing on three areas of graphic design-identity design, poster design, and web design-the study conducted literature reviews and expert surveys to evaluate the applicability and effectiveness of each AI tool across different design process stages. The results indicate that DALL·E excelled in stage 2 (idea generation), while Midjourney performed better in stage 3 (design prototyping) and stage 4 (design selection and refinement). Furthermore, Firefly, while initially showing lower performance in prompt creation, demonstrated gradual improvement as the process progressed, yielding comparatively positive results in tasks involving reference-based design refinement. This study provides valuable guidance for designers in selecting appropriate AI tools for their work and is expected to serve as foundational research for effectively integrating AI technology into the design process.

본 연구는 미해결 원가족 이슈를 가진 여성 내담자 4명을 대상으로 생성형 AI 이미지를 활용한 내러티브 상담 경험을 분석한 다중 사례연구이다. 생성형 AI로는 DALL-E 3를 사용하여 미해결 원가족 이슈와 선호하는 이야기를 AI 이미지로 시각화했고, 이를 디지털 매체로 내러티브 상담을 진행했다. 이후 반구조화 심층 인터뷰와 포커스 그룹 인터뷰를 실시하였다. 질적 분석 결과 첫째, ‘AI와의 만남’ 영역에서 AI는 상담의 숨은 조력자로 작용하여 연구참여자들이 미해결 원가족 이슈를 직면해 삶의 주도권 다시 찾기에 도움을 주었다. 둘째, ‘담론해체 이야기’ 영역에서 참여자들의 원가족 이해와 수용력이 증대되었다. 셋째, ‘선호하는 이야기’에서 현 가족 관계가 개선되면서 담론에서 벗어나 자기수용과 자기 돌봄 능력이 향상되는 효과가 있었다. 결론적으로, 본 연구는 가족치료 분야에서 생성형 AI 이미지를 활용한 내러티브 상담의 가능성을 탐구하고 그 의의를 논의하였다.

Objectives: This study is a multiple case study analyzing the counseling experiences of four women with unresolved family-of-origin issues through narrative therapy using AI image generation. Methods: DALL-E3 was used to convert family-of-origin issues and expectations into images, which were then used as tools for counseling. After completing five sessions of counseling, semi-structured interviews and focus group interviews were conducted. Results: The findings from the qualitative analyses were as follows. First, in the domain of “encountering AI”, AI was found to be an effective tool for helping participants to confront unresolved family-of-origin issues, overcome sadness and regain confidence and agency. Second, in the domain of “dissolution of discourse”, participants' understanding and acceptance of their family-of-origin increased. Third, in the domain of “alternative story”, participants’ current family relationships were improved, and their self-care/self-acceptance abilities were expanded. Conclusion: The potential use of image-generating AI in the family therapy field was discussed.

연구 목적: 본 연구는 GPT 기반 멀티모달 AI를 활용하여 시각장애인 학습자의 교과 학습 자료 내 시각 정보를 텍스트 기반 이미지 해설로 자동 변환하는 시스템을 고안하고, 그 품질과 적용 가능성을 실무 전문가를 통해 검증하고자 하였다. 연구 방법: 국내외 이미지 해설 가이드라인을 기반으로 GPT 확장 애플리케이션을 개발하였으며, 초·중학교 교과학습 자료에서 다양한 유형의 시각 자료를 수집하여 변환하였다. 생성된 이미지 해설 결과물에 대해 대체자료 제작 전문가 13인을 대상으로 품질 평가, 튜링 테스트, 만족도 및 사회적 타당도 평가를 실시하였다. 연구 결과: 품질 평가는 전반적으로 높은 수준을 보였으나 이미지 유형에 따라 차이가 나타났으며, 특히 그래프·차트 유형에서 상대적으로 낮은 평가가 확인되었다. 튜링 테스트에서는 AI가 생성한 해설이 가이드라인 예시와 유사한 수준으로 평가되어 일정 수준의 전문성을 확보한 것으로 나타났다. 만족도 및 사회적 타당도 또한 전반적으로 긍정적인 평가를 받았다. 논의 및 결론: 멀티모달 AI 기반 이미지 해설 생성 시스템이 시각장애인 학습자를 위한 대체자료 제작의 효율성과 품질을 향상시킬 가능성을 확인하였다. 다만, 정보량 조절, 유형별 품질 편차, 사용자 인지 수준 반영 등에서 기술적 보완이 요구된다. 후속 연구에서는 실무 현장 적용, 사용자 중심 평가 확대, 적응형 생성 시스템 개발을 통한 기술 고도화가 필요하다.

Purpose: This study developed a GPT-based multimodal AI system to automatically generate text-based image descriptions for students with visual impairments, and evaluated its quality and applicability. Methods: A custom GPT-powered application was built to convert educational visual materials into text descriptions, adhering to international image description guidelines. The outputs were evaluated by 13 experts through a three-part assessment: a quality evaluation, a Turing test to gauge human-likeness, and satisfaction ratings. Results: The AI system consistently produced high-quality descriptions that were often indistinguishable from those created by humans, as evidenced by the Turing test results and high expert ratings on satisfaction and social validity. However, the system struggled with the complexity of graphs and charts. Discussion and Conclusion: GPT-based multimodal AI can be a promising, efficient solution for generating high-quality image descriptions for visually impaired students. Future work should refine information density control and adapt outputs to user cognitive needs for broader real-world application.