turing's test vs the moral turing test by diane proudfoot
Philosophies, 2022
In this paper I argue that Turing proposed a new approach to the concept of thinking, based on hi... more In this paper I argue that Turing proposed a new approach to the concept of thinking, based on his claim that intelligence is an 'emotional concept'; and that the response-dependence interpretation of Turing's 'criterion for "thinking"' is a better fit with his writings than orthodox interpretations. The aim of this paper is to clarify the response-dependence interpretation, by addressing such questions as: What did Turing mean by the expression 'emotional'? Is Turing's criterion subjective? Are 'emotional' judgements decided by social consensus? Turing's take on these issues impacts current philosophical debates on response-dependent concepts and on the nature of artificial intelligence.
Artificial Intelligence, 2011
The widespread tendency, even within AI, to anthropomorphize machines makes it easier to convince... more The widespread tendency, even within AI, to anthropomorphize machines makes it easier to convince us of their intelligence. How can any putative demonstration of intelligence in machines be trusted if the AI researcher readily succumbs to make-believe? This is (what I shall call) the forensic problem of anthropomorphism. I argue that the Turing test provides a solution. This paper illustrates the phenomenon of misplaced anthropomorphism and presents a new perspective on Turing's imitation game. It also examines the role of the Turing test in relation to the current dispute between human-level AI and 'mindless intelligence'.
Turing's test vs the Moral Turing Test
Philosophy & Technology, 2024
Given actual autonomous systems with capacities for harm and the public’s apparent willingness to... more Given actual autonomous systems with capacities for harm and the public’s apparent willingness to take moral advice from large language models (LLMs), Einar Duenger Bohn’s (2024) renewed discussion of the Moral Turing Test (MTT) is timely. Bohn’s aim is to defend an unequivocally behavioural test. In this paper, I argue against this direction. Interpreted as testing mere behaviour, the Turing test is a poor test of either intelligence or moral agency, and neither Bohn’s version of the test nor Allen, Varner and Zinser’s influential (2000) version avoids these problems. Also, the MTT’s advantages as advertised by Bohn do not hold up: it is an open empirical question whether embracing a merely behavioural MTT will significantly reduce the challenges in building a computer to pass the test. This issue aside, I argue that Turing’s actual test is superior to current AI benchmarks and to Bohn’s version of the test. A test of moral reasoning or agency in machines that is based on Turing’s actual test has advantages over current tests of moral reasoning in AI, including versions of the MTT. Such a test is an intriguing possibility still to be investigated.
The Turing Test
MIT Open Encyclopedia of Cognitive Science, 2024
Turing, Wittgenstein and the Sciences of the Mind. A critical Notice of Justin Leiber'An Invitation to Cognitive Science
International Journal of Social Robotics, 2014
Anthropomorphism is a phenomenon that describes the human tendency to see human-like shapes in th... more Anthropomorphism is a phenomenon that describes the human tendency to see human-like shapes in the environment. It has considerable consequences for people's choices and beliefs. With the increased presence of robots, it is important to investigate the optimal design for this technology. In this paper we discuss the potential benefits and challenges of building anthropomorphic robots, from both a philosophical perspective and from the viewpoint of empirical research in the fields of human-robot interaction and social psychology. We believe that this broad investigation of anthropomorphism will not only help us to understand the phenomenon better, but can also indicate solutions for facilitating the integration of human-like machines in the real world.
On Wittgenstein on Cognitive Science
Philosophy, 1997
Cognitive science is held, not only by its practitioners, to offer something distinctively new in... more Cognitive science is held, not only by its practitioners, to offer something distinctively new in the philosophy of mind. This novelty is seen as the product of two factors. First, philosophy of mind takes itself to have well and truly jettisoned the ‘old paradigm’, the theory of the mind as embodied soul, easily and completely known through introspection but not amenable to scientific inquiry. This is replaced by the ‘new paradigm’, the theory of mind as neurally-instantiated computational mechanism, relatively opaque to introspection and the proper subject of detailed empirical investigation. Second, in the constitutive disciplines of cognitive science (cognitive psychology, neuroscience, computer science etc.) we have for the first time the theoretical, experimental and technological resources to begin this investigation. My concern here is to show that, despite its scientific and philosophical sophistication, the new (computational) paradigm is in certain striking ways very simi...
Science, 1999
The Age of Spiritual Machines When Computers Exceed Human Intelligence. Ray Kurzweil. Viking, New... more The Age of Spiritual Machines When Computers Exceed Human Intelligence. Ray Kurzweil. Viking, New York, 1999. 400 pp. $25.95. ISBN 0-670-88217-8. Kurzweil predicts that the next century will witness the fusion of humans and machines, but the reviewer finds such predictions for the future undercut by shortcomings in the book's account of the present and recent past.
Facts about Artificial Intelligence
Science, 1999
Ray Kurzweil (Letters, Science 's Compass, 16 July, p. [339][1]) responds to my review ( Scie... more Ray Kurzweil (Letters, Science 's Compass, 16 July, p. [339][1]) responds to my review ( Science 's Compass, 30 Apr., p. [745][2]) of his The Age of Spiritual Machines (Viking, New York, 1999) as follows. 1. 1) My review “mires the reader in obscure and misleading factual objections.” Kurzweil attempts a history of computing; in history, facts matter. He challenges only one of my historical objections, concerning the UNIVAC computer. His book, in an entry labeled “1950,” says, “Eckert and Mauchley develop UNIVAC, the first commercially marketed computer. It is used to compile the results of the U.S. census” (p. 269). In fact UNIVAC was under more or less continuous development from 1947; it was not the first commercially marketed computer, nor was it operational until 1951. 2. 2) I “drag out old anti-artificial-intelligence (AI) arguments.” I do not. Rather, I hold that make-believe about basic conceptual issues, such as we find in Kurzweil's book, are hindering AI. 3. 3) I complain “about anthropomorphizing, but there is no harm….” In AI, anthropomorphizing leads to an emphasis on human qualities that are irrelevant to, and a distraction from, the real aims of AI. 4. 4) My review “ignores [the book's] salient arguments….” I do not detect any, only fantasy, Kurzweil's own “laws” of physics, unjustified assertions, and factual errors. His letter is no different. For example, Kurzweil insists that Wittgenstein's Tractatus is about the brain, supporting this with a fallacious argument. In fact, the Tractatus is a technical work of symbolic and philosophical logic and abstract metaphysics and has nothing to say about the brain. Moreover, when Wittgenstein later did discuss the brain, he denied precisely Kurzweil's argument, that to talk about “thinking” or “knowing” is to talk about brain activity. Kurzweil also says that “there is nothing to prevent these efforts [modest connectionist experiments] from scaling up to the entire human brain.” How could he, or anyone else, possibly know this, given the vast discrepancy in scale that is involved (there are perhaps as many as 1014 neurons in the human brain)? [1]: /lookup/doi/10.1126/science.285.5426.337g [2]: /lookup/doi/10.1126/science.284.5415.745
Anthropomorphism is a common phenomenon known already in ancient times. It is not a thing of the ... more Anthropomorphism is a common phenomenon known already in ancient times. It is not a thing of the past, but still has a profound impact on major aspects of our lives and on research in AI and HRI. Its importance in the field of HRI is emphasized by the hotly-discussed uncanny valley hypothesis. However, in spite of its popularity, the uncanny valley hypothesis lacks empirical evidence. In this paper we suggest that the community should stop trying to fit data to this hypothesis, but rather, based on the available evidence, start talking about the 'uncanny curve'. Moreover, we point out mistakes in the previous studies of the uncanny curve and strongly encourage exploring it in a real HRI for it to be really relevant. We suggest that understanding the opposite process of anthropomorphisation, known as dehumanization, can help to cross the uncanny bottom of the graph.
Alan Turing's Forgotten Ideas In Computer Science
Scientific American, 1999
Alan Turing's forgotten ideas in computer science. BJ Copeland, Diane Proudfoot Scientific A... more Alan Turing's forgotten ideas in computer science. BJ Copeland, Diane Proudfoot Scientific American 280:44, 98-103, 4/1999. A survey is presented of Alan Turing's virtually unknown work anticipating the current fields of 'connectionism', or neuron-like computational processes. ...
Papers by diane proudfoot
Ali Hossein Khani and Gary N. Kemp (eds), Wittgenstein and Other Philosophers, Volume II. London & New York: Routledge, 2025
Did Wittgenstein's 1939 lectures on the foundations of mathematics-the one certain set of interac... more Did Wittgenstein's 1939 lectures on the foundations of mathematics-the one certain set of interactions between Wittgenstein and Turing-shape Turing's philosophy of mind? I argue that Turing's and Wittgenstein's criteria for thinking are similar in significant respects, and I investigate possible origins of these similarities in the 1939 lectures. The scholarly literature on the lectures focuses on mathematics and mathematical logic, yet in these lectures Wittgenstein presented several of his general philosophical maxims-maxims with which Turing's criterion for thinking is in agreement. The case for the hypothesis that Wittgenstein influenced Turing's philosophical approach to the mind is circumstantial but plausible.
Esercizi Filosofici, 2024
In this paper we apply Anscombe's account of human linguistic practices and of intentional action... more In this paper we apply Anscombe's account of human linguistic practices and of intentional action in a novel way-to the acts, by members of the Ngāi Tahu tribe in colonial-era Aotearoa New Zealand, of writing Letters to the Editor of local and regional newspapers. We identify the salient contexts of those acts and then draw on Anscombe's work to identify intentional and moral actions that otherwise risk going unnoticed. Our analysis exemplifies Anscombe's and Wittgenstein's view that we can read off intentional action from behaviour in context and also yields a rich historical and conceptual account of the letter-writers. We argue that this approach has theoretical advantages for Indigenous and Postcolonial Studies.
Australasian Journal of Logic, 2018
According to Richard Routley, a comprehensive theory of fiction is impossible, since almost anyth... more According to Richard Routley, a comprehensive theory of fiction is impossible, since almost anything is in principle imaginable. In my view, Routley is right: for any purported logic of fiction, there will be actual or imaginable fictions that successfully counterexample the logic. Using the example of 'impossible' fictions, I test this claim against theories proposed by Routley's Meinongian contemporaries and also by Routley himself (for what he called 'esoteric' works of fiction) and his 21st century heirs. I argue that the phenomenon of impossible fictions challenges even today's modal Meinongians.
FILOZOFIA I NAUKA, 2023
We examine Turing's intriguing claim, made in the philosophy journal Mind, that he had created a ... more We examine Turing's intriguing claim, made in the philosophy journal Mind, that he had created a short computer program of such a nature that it would be impossible "to discover by observation sufficient about it to predict its future behaviour, and this within a reasonable time, say a thousand years" (Turing, 1950, p. 457). A program like this would naturally have cryptographic applications, and we explore how the program would most likely have functioned. Importantly, a myth has recently grown up around this program of Turing's, namely that it can be used as the basis of an argument-and was so used by Turing-to support the conclusion that it is impossible to infer a detailed mathematical description of the human brain within a practicable timescale. This alleged argument of Turing's has been dubbed "Turing's Wager" (Thwaites, Soltan, Wieser, Nimmo-Smith, 2017, p. 3) We demonstrate that this argument-in fact nowhere to be found in Turing's work-is worthless, since it commits a glaring logical fallacy. "Turing's Wager" gives no grounds for pessimism about the prospects for understanding and simulating the human brain.
The Journal of Philosophy, 2013
he Turing test is one of the philosophical foundations of Artificial Intelligence. In the sixty y... more he Turing test is one of the philosophical foundations of Artificial Intelligence. In the sixty years since Alan Turing's "Computing Machinery and Intelligence" appeared in Mind, there have been two widely accepted interpretations of the test-the canonical behaviorist and the rival inductive (or epistemic) accounts. These accounts are based on Turing's 1950 Mind paper, and few commentators know that Turing described two other versions of his imitation game. These versions are scarcely mentioned in the voluminous literature on the Turing test but are (I shall argue) essential to understanding the test. Turing described the first version in 1948, the year he moved to Manchester to run the Computing Machine Laboratory at the university. 1 Earlier, in June 1948, the world's first electronic stored-program digital computer, the Manchester "Baby", ran its first program in the Laboratory. During 1948-49, the Baby was expanded to become a much more substantial machine, and in May 1949 the Electronic Delay Storage Automatic Calculator (EDSAC) became the second electronic stored-program computer to function, at the Mathematical Laboratory of the University of Cambridge. Further developments at Manchester led to the first commercially available electronic digital computer, in February 1951. In this way, in the late 1940s and early 1950s, the abstract question whether machine intelligence is possible came to be focused on a particular form of machine. Could machinery like the Baby and the EDSAC, if given additional high-speed memory and enhanced processing speed, be said to think? To investigate this question, Turing required a "criterion for 'thinking'" 2 -the "imitation game" was to provide this. In a 1948 report on machine learning entitled "Intelligent Machinery", he * I am indebted to Jack Copeland for comments on an earlier draft of this paper. 1 On the Manchester computers, see further B.
Books by diane proudfoot
Philosophical Explorations of the Legacy of Alan Turing, 2017
In 1948 Turing claimed that the concept of intelligence is an "emotional concept". An emotional c... more In 1948 Turing claimed that the concept of intelligence is an "emotional concept". An emotional concept is a response-dependent concept and Turing's remarks in his 1948 and 1952 papers suggest a response-dependence approach to the concept of intelligence. On this view, whether or not an object is intelligent is determined, as Turing said, "as much by our own state of mind and training as by the properties of the object". His discussion of free will suggests a similar approach. Turing said, for example, that if a machine's program "results in its doing something interesting which we had not anticipated I should be inclined to say that the machine had originated something". This points to a new form of free will compatibilism, which I call response-dependence compatibilism and explore here.
Wittgenstein and Artificial Intelligence, Volume I , 2024
Uploads
turing's test vs the moral turing test by diane proudfoot
Papers by diane proudfoot
Books by diane proudfoot