Undergraduate thesis attempts to presents Dreyfus' critique of artificial intelligence, focusing on the latest version of Hubert Dreyfus's book What Computers Still Can't Do: A Critique of Artificial Intelligence. Thesis first presents a definition of artificial intelligence, followed by a brief history of artificial intelligence, where two historical approaches to artificial intelligence are presented. This is followed by a presentation of four artificial intelligence assumptions that Dreyfus criticized for unjustified optimism. Assumptions follow each other from biological, through psychological and epistemological, to ontological. As part of the diploma thesis, Dreyfus' alternatives to traditional assumptions are also presented. The first is dedicated to emphasizing the role of the human body in intelligent behavior, the second to the phenomenon of orderly action without reference to the rules, and the last to the meaning of the situation as a function of human needs. Finally, I turn to the importance of criticism today, in the context of the relationship between the symbolic and connectionist approaches in artificial intelligence and the still-relevant problem of context today.
|
