인공지능 윤리에서 투명성의 원칙 고찰

초록

In this paper, I examined the principle of transparency, which is newly emphasized in the digital age, among various ethical principles. The principle of transparency is an ethical principle that is increasingly important as AI develops due to the black boxing of AI algorithms. First, I examined the relationship between ethical principles and the principle of transparency as conditions for trustworthy AI. Next, I looked at the singularity problem and the ‘heuristics of fear’. Third, I divided the subjects of transparency into humans and AI. In addition, I suggested the direction of transparency practice. In conclusion, I argued that AI stakeholders should practice the principle of transparency to prevent misuse, design errors, and unintended side effects of AI.

키워드