人工智能时代的价值审度与伦理调适

Value Reflection and Ethical Adjustment in the Era of Artificial Intelligence

人工智能体的拟主体性赋予了人工智能特有的拟伦理角色。通过对此拟伦理角色的分析,可以看到人工智能的价值审度与伦理调适的基本路径应为负责任的创新和主体权利保护。立足对合成智能与人造劳动者的价值审度,可提出寻求算法决策与算法权力的公正性,呼唤更加透明、可理解和可追责的智能系统,反思智能化的合理性及其与人的存在价值的冲突等价值诉求。应从有限自主与交互智能体概念出发,进一步展开对人工智能的价值校准和伦理调适,这不仅有助于把握人工智能伦理研究与相关伦理规范和标准的内涵,还可揭示"科技—产业—伦理"异质性实践所应有的明智。

The quasi-subiectivity of artificial intelligence agents gives artllmlai mtemgence a umque quasi ethical role. Through analysis of the quasi ethical role o{ the agents, we can see that the basic ap- proach of value reflection and ethical adjustment on the artificial intelligence should be responsible in- novation and the protection of subject rights. Based on the evaluation of the value o{ synthetic intelli- gence and artificial laborers, it is necessary to put forward some value pursuits, which include seeking fairness of algorithmic decision and algorithm power, calling for a more transparent, understandable and accountable intelligent system, reflecting on the confliction between the rationality of intelligence and the value of human existence. What is more, we should introduce the value alignment and ethical adjustment of AI based on the concept of limited autonomy and interactive agent, so as to facilitate the analyzing of the meaning of artificial intelligence ethics and related ethical norms and standards, as well as the exploringof the necessary prudence in the heterogeneity practice of ＂science and technology-in- dustry-ethics＂.