德雷福斯与海德格尔式人工智能

Artificial Intelligence and Being-in-the-World:Dreyfus on Heideggerian Artificial Intelligence

自20世纪60年代以来,德雷福斯从海德格尔哲学出发对人工智能进行了深入批判,积极地影响了人工智能研究的发展。他指出,把人的心灵当作信息处理设备的做法完全基于传统形而上学的认识论和存在论假设,不能够成功模拟人的智能。心智活动既不能够被完全形式化,也不能够被孤立地、离散地处理。德雷福斯对人工智能研究的直接影响是促使不同阶段的“海德格尔式人工智能”产生。但即便是这种试图模拟海德格尔“在-世界-中-存在”模式的人工智能仍不能突破框架问题。正如德雷福斯对麦克道尔的挑战所展示出的,人的智性活动特有的整体性、直觉性把握的特征有别于表征性、概念性的处理模式。海德格尔对人的存在样式的深刻剖析将继续为人工智能的发展提供根本的思想资源和参照系。

Hubert Dreyfus criticism of artificial intelligence based on Heidegger's philosophy has positively influenced the development of AI research since the 1960s.According to Dreyfus,considering the human mind as a type of information processer presupposes unsustainable traditional metaphysical and ontological hypotheses that are unable to adequately account for the ways in which human intelligence can be simulated,since human intelligence can neither be fully formalized nor dealt with in its discrete isolation.The direct impact of Dreyfus AI research led to the generation of different stages of research on AI inspired by Heideggerian philosophy.This study examines the ways in which even the kind of artificial intelligence which attempts to simulate Heidegger s“Being-in-the-world”,cannot break through the frame problem.As Dreyfus challenge to McDowell shows,human intellectual activities share the characteristics of wholeness and intuition,which make them quite different from representational and conceptual processing patterns.For these and other reasons discussed below,Heidegger's profound analysis of human existence will necessarily continue to provide fundamental thought resources and frames of reference for the development of artificial intelligence.