“机器意识”何以可能——人工智能时代“机器不能被骗”立场之坚守  

How is Machine Consciousness Possible——Sticking to the Position of“Machines Can’t Be Cheated”in the Era of Artificial Intelligence

作　　者：陈俊秀 李立丰[1] CHEN Junxiu;Li Lifeng(Law school, Jilin University, Changchun 130012, China)

机构地区：[1]吉林大学法学院,吉林长春130012

出　　处：《大连理工大学学报（社会科学版）》2020年第6期108-114,共7页Journal of Dalian University of Technology(Social Sciences)

基　　金：国家社会科学基金青年项目:“我国死刑案件审理过程中民意的拟制与导入机制研究”(13CFX060)。

摘　　要：传统刑法理论囿于机器不存在意识和认知能力,无法陷入“认识错误”而处分财物或财产性利益的事实层面,得出“机器不能被骗”的刑法命题。以形式化算法为基础的人工智能无法充分模拟以“生物算法”为基础的人类意识,“机器意识”在理论上的可能性和技术上的可行性均存有疑问。机器意识之实现与验证于目前而言无法证成,其将来实现之可能性也无法证伪,“机器不能被骗”理论毋宁视为从人类理性、人类尊严出发,降低人类未来生存风险所必须遵循的前提假定。将人工智能定位为法律的客体,排除其拥有独立于人的意识。纾解“机器不能被骗”之法教义学阐释路径:在“人机关系”中,机器之管理者被骗,即当行为符合人工智能程序的预设条件时,机器所作出的反馈性行为应当视为机器管理者的思想表达;机器处于“占有辅助者”地位,从属于上位者,通过事先内置特定的指令程序以实现机器管理者的意志,属于事实行为,而并非机器的“法律行为”;综合考量行为人的欺骗行为与机器预设同意范围,若行为突破管理者事先预设的同意条件范畴,不能认定为机器之管理者被骗。The traditional theory of criminal law is limited to the fact that the machine does not have consciousness and cognitive ability,and cannot fall into the“cognitive error”or dispose of property or property interests,so it comes to the criminal proposition that“the machine cannot be cheated”.Artificial intelligence based on formal algorithm cannot fully simulate human consciousness based on“biological algorithm”,and the theoretical possibility and technical feasibility of“machine consciousness”are questionable.At present,the realization and verification of machine consciousness can’t be proved,but the possibility of its realization in the future can’t be falsified.The theory of“machine can’t be deceived”is rather regarded as the premise that must be followed in order to reduce the risk of human survival in the future from the perspective of human rationality and human dignity.Artificial intelligence is defined as the object of law,excluding its consciousness as being independent of human beings.The doctrinal explanation to relieve“machine can not be deceived”is as follows:In“man-machine relationship”,the manager of the machine is cheated,that is,when the behavior meets the preset conditions of artificial intelligence program,the feedback behavior made by the machine should be regarded as the thought expression of the manager of the machine;As an“occupying assistant,”The machine is,subordinate to the upper level,using the built-in specific instruction program in order to realize the will of the machine manager;It is a factual act rather than a“legal act”of the machine;The deceptive act of the actor and the pre-set consent scope of the machine should be considered comprehensively.If the act breaks through the pre-set consent condition scope of the manager,it cannot be deemed that the manager of the machine is cheated.

关 键 词：人工智能 机器意识 诈骗罪 机器不能被骗 教义学阐释 

分 类 号：D924.34[政治法律—刑法学]