earticle

영어

As Artificial Intelligence (AI) becomes increasingly prevalent in everyday life, concerns about AI ethics and privacy are growing. One such concern is the unauthorized use of AI chatbots through a practice known as "jailbreaking." Jailbreaking refers to the act of bypassing software restrictions on chatbots, enabling users to use them in unauthorized ways. Our study aims to explore the motivations behind jailbreaking and its implications for ethical governance in the context of AI. We will employ a mixed-method approach, combining topic modeling and regression analysis to extract and analyze data from online forums related to jailbreaking. Our study has both academic and practical implications. The academic implications of this study include enhancing our understanding of the ethical and privacy implications of jailbreaking. The practical implications include informing organizations and policymakers of potential vulnerabilities in their AI chatbot systems and proactively preventing possible jailbreaking to protect user privacy and maintain ethical standards.