不可否定的藝術性? AI生成繪圖之價值評估歷程研究;Is Artistic Value Undeniable? A Study on the Evaluation Process of AI-generated Images

NCU Institutional Repository > 管理學院 > 企業管理研究所 > 博碩士論文 > Item 987654321/97543

請使用永久網址來引用或連結此文件: https://ir.lib.ncu.edu.tw/handle/987654321/97543

題名:	不可否定的藝術性? AI生成繪圖之價值評估歷程研究;Is Artistic Value Undeniable? A Study on the Evaluation Process of AI-generated Images
作者:	曾冠瑋;Tseng, Kuan-Wei
貢獻者:	企業管理學系
關鍵詞:	AI生成圖像;擬人化感知;美感經驗;美感情緒;價值評估;感知一致性
日期:	2025-07-15
上傳時間:	2025-10-17 11:30:25 (UTC+8)
出版者:	國立中央大學
摘要:	隨著生成式人工智慧（Generative AI）的迅速發展，AI圖像生成系統（text-to-image AI generation systems）逐漸普遍應用於藝術領域，引發大眾對其藝術性、價值等議題的關注與討論。為深入理解觀者如何評估AI生成圖像作品的價值，本研究建構整合擬人化感知、藝術反應與價值評估之模型，以探討觀者評價之心理歷程。研究中以自主性與經驗情感作為觀者對AI繪圖機器人之擬人化感知，將藝術反應區分為認知層面的美感經驗與情感層面的美感情緒，並引入感知一致性作為干擾變項，以檢視其對價值評估機制之調節效果。本研究透過網路問卷蒐集資料，共回收有效樣本495份，以結構方程模型進行假說分析。實證結果顯示，經驗情感透過美感情緒進一步影響商品價值，為本研究之最強路徑，顯示觀者對AI模擬情感能力的感知與本身的情感投入，是評估商品價值的關鍵因素。此外，僅情感層面的美感情緒對價值評估具正向影響，突顯觀者更重視與作品的情感連結而非視覺呈現。另一方面，感知一致性僅對藝術價值評價產生正向顯著的干擾作用，代表當觀者認為AI與人工創作在投入與成果上相當時，會產生價值一致性的心理認同，進而強化對藝術價值的評價。 ;With the rapid advancement of generative artificial intelligence (Generative AI), text-to-image AI generation systems have become increasingly prevalent in the field of art, sparking widespread public discussion regarding their artistic legitimacy and value. To gain a deeper understanding of how viewers evaluate the value of AI-generated artworks, this study proposes an integrated model that incorporates anthropomorphized perception, artistic response, and value assessment, aiming to explore the psychological processes underlying viewers′ evaluations. In this model, agency and experience represent viewers’ anthropomorphized perception of AI text-to-image AI generation systems. Artistic response is conceptualized along two dimensions: aesthetic experience, which reflects cognitive engagement, and aesthetic emotion, which reflects emotional engagement. Perceived fairness is introduced as a moderating variable to examine its influence on the value evaluation mechanism. A total of 495 valid responses were collected through an online survey, and the hypotheses were tested using structural equation modeling (SEM). The empirical results show that experience influences commodity value through aesthetic emotion, constituting the strongest path in the model. This indicates that viewers’ perception of AI’s emotional capabilities and their own emotional involvement are key factors in the evaluation of commodity value. Moreover, only aesthetic emotion has a significant positive effect on value assessment, suggesting that viewers place greater importance on emotional connection with the artwork than on its visual presentation. In addition, perceived fairness exerts a significant moderating effect only on the evaluation of artistic value, indicating that when viewers perceive AI’s effort and outcome as comparable to those of human creators, their recognition of artistic value is strengthened.
顯示於類別:	[企業管理研究所] 博碩士論文

文件中的檔案:

檔案	描述	大小	格式	瀏覽次數
index.html		0Kb	HTML	11	檢視/開啟

在NCUIR中所有的資料項目都受到原著作權保護.

社群 sharing

資料載入中.....