As artificial intelligence gradually replaces human decision-making in areas such as recruitment, credit approval, and criminal justice, the tension between the efficiency gains it brings and the potential risk of bias becomes increasingly pronounced. In your view, when algorithms may inadvertently entrench social inequality, should we restrict the autonomous judgment of AI in critical public affairs? Please articulate your position, supported by concrete examples and logical reasoning.