현재 위치
Home
Culture Convergence (CC)
The Limits of AI-Generated Creativity: Understanding the Challenges of User Intent Reflection
-
- 발행기관
- 국제문화기술진흥원 바로가기
-
- 간행물
- International Journal of Advanced Culture Technology(IJACT) KCI 등재 바로가기
-
- 통권
- Volume 13 Number 1 (2025.03)바로가기
-
- 페이지
- pp.312-321
-
- 저자
-
- 언어
- 영어(ENG)
-
- URL
- https://www.earticle.net/Article/A465929
※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.
원문정보
초록
- 영어
- AI-driven image generation has transformed digital creativity but remains fundamentally limited in adhering to explicit design rules. Unlike traditional creative software, which incorporates structured validation and editing processes, AI-driven creative systems operate without real-time analytical refinement, leading to significant limitations in precision and accuracy. This study identifies a key issue: the disconnect between image analysis AI and image generation AI, resulting in persistent errors in numerical accuracy, spatial consistency, and stylistic adherence. Through a systematic evaluation of DALL·E 3, MidJourney, and Stable Diffusion 3.5, we demonstrate that even after 100 iterations, AI-generated outputs exhibited critical errors in fundamental concepts such as facial features (e.g., eyes, nose, and mouth) and numerical constraints (e.g., generating exactly three teeth), revealing AI’s inability to process structured and quantitative instructions reliably. To address these challenges, this study proposes a bidirectional feedback mechanism that integrates analytical validation layers into generative models, enabling real-time refinement. By establishing an AI interoperability framework, we bridge the gap between generative AI and analytical AI, allowing for real-time validation and iterative improvement of generated outputs. Additionally, we emphasize dataset optimization for geometric and minimalist structures and advocate for human-AI collaborative refinement to improve output accuracy. These findings highlight the necessity of structured analytical feedback in AI-driven creative systems, which we expect could contribute to the development of more adaptive, user-controlled design workflows.
목차
Abstract
1. INTRODUCTION
2. RESEARCH METHOD
2.1 Selection Criteria for AI Tools
2.2 Evaluation Criteria
2.3 Data Collection and Analysis Process
3. RESULTS AND DISCUSSION
3.1 Limitations in Text-Based Command Interpretation
3.2 Common Failures Across AI Tools
3.3 Dataset Dependency and Style Limitations
3.4 A Fundamental Gap Between Expectation and Reality
4. Conclusion
4.1 Proposed Solutions and Future Directions
4.2 Research Limitations and Future Work
ACKNOWLEDGEMENT
REFERENCES
키워드
AI-Assisted Creativity User Intent Recognition in AI Generation Misinterpretation in AI Outputs Creative Limitations of AI AI Interoperability in Text-to-Image Systems저자
간행물 정보
발행기관
-
- 발행기관명
- 국제문화기술진흥원 [The International Promotion Agency of Culture Technology]
- 설립연도
- 2009
- 분야
- 공학>공학일반
- 소개
- 본 진흥원은 문화기술(Culture Technology) 관련 산·학·연·관으로 구성된 비영리 단체이다. 문화기술(CT)은 정보통신기술(ICT), 문화적 사고 기반의 예술, 인문학, 디자인, 사회과학기술이 접목된 신융합기술(New Convergence Technology, NCT)로 정의한다. 인간의 삶의 질을 향상시키고, 진보된 방향으로 변화시키고, 문화기술 관련 분야의 학술 및 기술의 발전과 진흥에 공헌하기 위하여, 제3조의 필요한 사업을 행함을 그 목적으로 한다.
간행물
-
- 간행물명
- International Journal of Advanced Culture Technology(IJACT)
- 간기
- 계간
- pISSN
- 2288-7202
- eISSN
- 2288-7318
- 수록기간
- 2013~2025
- 등재여부
- KCI 등재
- 십진분류
- KDC 600 DDC 700
이 권호 내 다른 논문 / International Journal of Advanced Culture Technology(IJACT) Volume 13 Number 1
피인용수 : 0건 (자료제공 : 네이버학술정보)
함께 이용한 논문 이 논문을 다운로드한 분들이 이용한 다른 논문입니다.
출처 : 네이버학술정보
0개의 논문이 장바구니에 담겼습니다.