生成式人工智慧有關兒少性剝削製品問題

Child Sexual Exploitation Material Created by Generative AI: Issues and Challenges

生成式人工智慧的發展為數位內容創作帶來了重大變革，特別是影像與音訊的生成技術。然而隨著生成式人工智慧技術的成熟，社會亦面臨新的倫理與法律挑戰，人工智慧生成的兒少性剝削製品問題即為其一。近年來不斷出現人工智慧生成的兒少性剝削製品，包括利用深偽技術及生成模型生成的高度擬真的性圖像或影像等，嚴重侵害兒少的身心健康及隱私權。根據美國國家失蹤與受虐兒少援助中心及多項國際研究顯示，人工智慧生成技術不僅大幅增加兒少性剝削圖像的生產數量，更增加警察執法部門辨識和追蹤被害者及加害者的難度。本研究旨在分析生成式人工智慧技術被濫用的現況及相關問題，並進一步探討國際社會於法制層面與實務層面的回應措施與因應策略，期能作為我國未來法制發展與政策規劃之參考。

The rapid advancement of generative AI technology has significantly transformed digital content creation, particularly in image and audio generation. However, alongside technological maturation, society is increasingly confronted with emergent ethical and legal challenges, notably the proliferation of generative AI-produced child sexual exploitation material. Recent years have witnessed a surge in highly realistic AI-generated child sexual exploitation material, facilitated by deepfake technology and generative models, resulting in substantial harm to minors’physical and psychological well-being and a serious violation of their privacy rights. According to reports by the National Center for Missing & Exploited Children (NCMEC) and various international studies, generative AI technologies have significantly escalated the volume of child sexual exploitation material production, complicating law enforcement agencies' efforts to identify and track victims and offenders. This paper examines the current risks and challenges posed by the misuse of generative AI technology, explores international legal frameworks and practical response strategies addressing this issue, and provides recommendations for informing future legal developments and policy planning in the national context.