试论人工智能的伦理责任

On the Ethical Responsibility of Artificial Intelligence

人工智能在社会中的作用越来越突出,但同时也对我们现有的社会伦理责任体系造成很大的冲击,很容易形成责任空白问题。新工具主义主张将智能机器视为人类的奴隶,机器伦理的支持者倾向于赋予智能机器以道德主体地位,这两种方案在处理人工智能的伦理责任问题时都面临众多困难。相比较而言,混合方案更为合理。从整体论的视角出发,混合方案主张将道德主体扩展至他者以及非人类的实体。通过修改扩展的主体这一概念,智能机器导致的责任问题可以指派给所有与智能机器相关的责任方。

It is true that artificial intelligence is playing an increasingly important role in today’s society, yet it also challenges the current social ethical system by triggering the question: who should be responsible for the accidents involving intelligent machines? The new instrumentalism claims that machines should be regarded as slaves. The supporters of machine ethics hold that intelligent machines should be regarded as moral agents. However, both stances are faced with many challenging problems. Comparatively speaking, the hybrid approach is more rational in that it takes the holistic perspective, arguing that the status of moral agent should be extended to nonhuman entities. Following this rationale, all agents related to intelligent machines should be assigned for the joint responsibility in events involving intelligent machines.