Eugen Dimant examines how people increasingly outsource moral reasoning to AI systems, identifying three distinct risks. “Cognitive surrender” occurs when users adopt AI-generated answers without critical evaluation, potentially amplifying existing biases. Delegating decisions to AI can also increase dishonest behavior, as the technology appears to absorb moral responsibility from the user, while chronic over-reliance risks a form of “moral deskilling” in which ethical judgment atrophies through disuse.
