As artificial intelligence (AI) becomes increasingly integrated into decision-making processes, understanding the psychological factors that shape human willingness to delegate tasks to AI is critical. This study explores how perceived losses—such as diminished control, accountability, or personal value—affect individuals’ autonomy in decision-making when interacting with AI systems. Through a series of behavioral experiments and surveys, findings reveal that higher perceptions of loss significantly reduce the likelihood of AI delegation, even when efficiency or accuracy is improved. The results also indicate that trust in AI and perceived competence partially mediate this relationship. These insights have implications for AI interface design, organizational decision policies, and ethical considerations in human-AI collaboration.