Dans son célèbre article de 1950 «Computing Machinery and Intelligence», Alan Turing suggère qu'un critère d'intelligence pour une machine serait qu'elle puisse se faire passer pour un interlocuteur humain. Ce test, appelé «Test de Turing», a été critiqué par John Searle, qui a remis en cause sa pertinence en s'appuyant sur une expérience mentale connue sous le nom de l'argument de la chambre chinoise. L'expérience tend à montrer que l'intelligence artificielle au sens propre du terme est impossible, parce que les (...) symboles manipulés par les programmes de prétendue «intelligence artificielle» sont dépourvus de signification. L'intentionalité propre à certains êtres vivants peut seule faire naître le sens. L'imitation d'un comportement intentionnel ne le peut pas. Le test de Turing serait done insatisfaisant, parce qu'il ne permettrait pas de distinguer les systèmes intentionnels de systèmes qui ne le sont pas, et que par conséquent il ne permettrait pas de distinguer l'intelligence authentique de la simulation de l'intelligence. (shrink)

What happens to philosophy when philosophical activities migrate to the AI lab? My article explores the philosophical work that has gone into the machine simulations of language and understanding after Alan Turing. The early experiments by AI practitioners such as Karen Spärck Jones, Richard Richens, Yorick Wilks, and others at the Cambridge Language Research Unit (CLRU) led to the creation of the machine interlingua, semantic networks, and other technological innovations central to the development of AI in the 1950s–1970s. I attempt (...) to show how, in the midst of their computational work, the CLRU pioneers engaged with Ludwig Wittgenstein, Ferdinand de Saussure, Rudolf Carnap, and other philosophers and developed startling new ways of formulating fundamental questions about language and human understanding. More significantly, their philosophical activities on the machine present an inclusive and culturally diverse picture of the world that profoundly negates the ethnocentric metaphysics of human-machine conundrums that John Searle and his critics represent in the Chinese Room debate. The familiar legacy of that debate has long distorted the narrative of AI origins through its simultaneous reiteration and repudiation of the Turing test. My study seeks to clarify those origins, but my primary goal is to demonstrate what it is like to practice philosophy on the machine and how the critique of metaphysics is made possible in the AI lab. (shrink)

We frame the question of what kind of subjective experience a brain simulation would have in contrast to a biological brain. We discuss the brain prosthesis thought experiment. Then, we identify finer questions relating to the original inquiry, and set out to answer them moving forward from both a general physicalist perspective, and pan-experientialism. We propose that the brain simulation is likely to have subjective experience, however, it may differ significantly from human experience. Additionally, we discuss the relevance of quantum (...) properties, digital physics, theory of relativity, and information theory to the question. (shrink)