理解生成式AI:对一个互联网发展史上标志性节点的审视  

Understanding Generative Al:An Examination of a Milestone in the History of the Internet'sDevelopment

作　　者：喻国明 刘彧晗 Yu Guoming;Liu Yuhan

机构地区：[1]北京师范大学新闻传播学院,北京100875

出　　处：《传媒观察》2023年第9期36-44,共9页Media Observer

摘　　要：ChatGPT问世掀起生成式AI研究热潮。本文从复杂性范式出发,依循“是什么-会怎样-应如何”的逻辑分析生成式AI的理解路径。生成式AI是实现传播理性与非理性要素交织的新媒介技术,将内容网络升维成更具开放性的复杂巨系统。从社会影响来看,生成式AI作为智能主体和智能工具,可通过“替代”与“增强”人类脑力的方式,促进人类非理性逻辑与机器理性逻辑交织并深入社会表达之中,使普罗大众得以跨越传播“能力沟”并实现平均水平的智力增强,进一步打破精英宰制的社会并迈入“常人政治”的未来新社会。在治理方面,应摒弃机械控制论思路,以其道德伦理和算法伦理为原则,形成抓大放小的复杂性治理思路。面对未来发展,为避免陷入科林格里奇困境,应沿智能试错方案的思路,允许通过技术小规模试错来厘清发展路径,相关研究也应更关注生成式AI具体应用层的议题。The emergence of ChatGPT has set off a boom in generative AI research.This paper explains the understanding path of generative AI from the complexity paradigm,following the logic of what it is-what it will be-how it should be.Generative AI is a new media technology that realizes the interweaving of rational and irrational communicational elements,and upgrades the content network into a more open and complex system.Generative AI,as an intelligent subject and an intelligent tool,can realize the effects of intelligent substitution and intelligent augmentation.For example,the general public can cross the ability gap and realize the average level of intellectual enhancement,and further break down the society dominated by the elites and enter the politics era of ordinary people.We should abandon the idea of mechanical cybernetics,and form a complexity governance based on the principles of moral ethics and algorithmic ethics.In order to avoid falling into the Colingridge dilemma,we should follow the idea of intelligent trial-and-error program to clarify the development path,and the research should also pay more attention to the specific application layer of generative AI.

关 键 词：生成式AI 复杂性范式 非理性因素 适应性主体 涌现 再组织化 

分 类 号：G206[文化科学—传播学]