<![CDATA[The development of artificial intelligence (deepfake) technology has been widely misused to manipulate a person's face into that of another person, potentially leading to its misuse in crimes such as fraud, misuse of personal data, extortion, pornography, and political sabotage. Based on this, this study aims to determine how civil liability applies to the misuse of deepfake technology and what challenges and solutions exist in applying civil liability to those who misuse deepfake technology. Using a normative research approach combined with conceptual and legal frameworks, this study examines all laws and regulations related to civil liability for the misuse of deepfake technology. The results of this study indicate that both providers and users of deepfake platforms can be held liable for negligence or errors that cause harm, as per the concept of unlawful acts regulated in Articles 1365 and 1367 of the Civil Code. However, the implementation of civil liability for deepfake misuse still faces significant complexities. These challenges stem from technical aspects, such as the limitations of detection model generalization and the high realism of deepfake content driven by deep learning technology, as well as the ease of access to software. Enforcement of the law faces obstacles in identifying perpetrators, cross-border jurisdictional issues, and difficulties in quantifying immaterial losses. Additionally, there is a specific legal vacuum that explicitly regulates deepfakes, making existing legal regulations general and not yet adaptive to the characteristics of artificial intelligence-based manipulation.]]>
