人工智能领域伦理主体agent(行为体)的追本溯源  

The Origin of the Ethical Subject AGENT in the Field of Artificial Intelligence

作　　者：寿步[1] SHOU Bu(School of Law,Shanghai Jiao Tong University,Shanghai,200030)

机构地区：[1]上海交通大学法学院,上海200030

出　　处：《自然辩证法通讯》2023年第12期10-17,共8页Journal of Dialectics of Nature

摘　　要：伦理学与AI的交叉研究所涉及的伦理主体指向AI中的agent。但AI中agent的中文译名尚未形成共识,这限制了伦理学与AI交叉研究的深入。回顾AI中出现agent的时代背景,考察AI经典英文论著中的agent论述,可以确定agent是具有行动能力的存在,是行动或行使权力者,应译为“行为体”(doer或actor)。应当区分“行为体”和“智能行为体”。没有一个“行为体”自身是有智能的;只有出现涌现性的“行为体社会”才可能是智能的。此项正名为伦理学研究AI中agent的行为给出了循名责实的线索、预留了名正言顺的空间。The ethical subject involved in the intersection of ethics and Artificial Intelligence points to the AGENT in AI.There is no consensus on the Chinese translation of the AGENT in AI,which limits the depth of research on interdisciplinary of ethics and AI.Reviewing the era background of the emergence of AGENT in AI,examining the descriptions of AGENT in classic AI works in English,it can be determined that an AGENT is a being with the capacity to act,the one that acts or exerts power,and should be translated as xingweiti(doer or actor).A distinction should be made between an AGENT and an intelligent AGENT.None of AGENTs is itself intelligent.Only an emergent“society of AGENTs”can be intelligent.This rectification is a clue to literally give the name,so as to study the action of AGENT in the interdisciplinary research of AI and ethics.

关 键 词：伦理主体 行为体 行为力 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]