Anthropomorphism and AI: Turings much misunderstood imitation game

Artificial Intelligence

Volumn 175, Issue 5-6, 2011, Pages 950-957

  Proudfoot, Diane ^a  

^a UNIVERSITY OF CANTERBURY   (New Zealand)

Author keywords

Anthropomorphism; Human level AI; Mindless intelligence; Turing test

Indexed keywords

ANTHROPOMORPHISM; HUMAN LEVELS; IMITATION GAMES; MINDLESS INTELLIGENCE; TURING TESTS;

ARTIFICIAL INTELLIGENCE;

EID: 79953183327     PISSN: 00043702     EISSN: None     Source Type: Journal    
DOI: 10.1016/j.artint.2011.01.006     Document Type: Review

References (110)

1. See, for example, N. Cassimatis, E.T. Mueller, and P.H. Winston, "Achieving human-level intelligence through integrated systems and research: introduction to this Special Issue", AI Magazine, vol. 27, no. 2, 2006.
2. also E.A. Feigenbaum, "Some Challenges and Grand Challenges for Computational Intelligence", Journal of the ACM, vol. 50, no. 1, 2003.
3. D.B. Lenat, "The voice of the turtle: whatever happened to AI?", AI Magazine, vol. 29, no. 2, 2008.
4. B. Goertzel, "Human-level artificial general intelligence and the possibility of a technological singularity: A reaction to Ray Kurzweil-s The Singularity Is Near, and Mc-Dermott-s critique of Kurzweil", Artificial Intelligence, vol. 171, 2007.s
5. J. McCarthy, "From Here to Human-Level AI", Artificial Intelligence, vol. 171, 2007.
6. "Formalizing common sense knowledge in mathematical logic", The Rutherford Journal: The New Zealand Journal for the History and Philosophy of Science and Technology, vol. 3, www.rutherfordjournal.org.
7. M. Minsky, P. Singh, and A. Sloman, "The St. Thomas common sense symposium: designing architectures for human-level intelligence", AI Magazine, vol. 25, no. 2, 2004.
8. L.A. Zadeh, "Toward Human Level Machine Intelligence-Is It Achievable? The Need for a Paradigm Shift", IEEE Computational Intelligence Magazine, vol. 3, no. 3, 2008.
9. J. McCarthy, "From Here to Human-Level AI", p. 1174. In the same issue Goertzel makes an analogous claim: -AGI should be the focus of a significant percentage of contemporary AI research . . . [and] dramatic progress on AGI in the near future is something that-s reasonably likely- ("Human-level artificial general intelligence and the possibility of a technological singularity: A reaction to Ray Kurzweil-s The Singularity Is Near, and McDermott-s critique of Kurzweil", p. 1163).
10. See, for example, D. McDermott, "Level-headed", Artificial Intelligence, vol. 171, 2007.
11. A. Sloman, "The well-designed young mathematician", Artificial Intelligence, vol. 172, 2008. Goertzel criticises the goal of human-level AI but is happier with AGI ("Human-level artificial general intelligence and the possibility of a technological singularity: A reaction to Ray Kurzweil-s The Singularity Is Near, and McDermott-s critique of Kurzweil"). Several of McDermott-s and Sloman-s objections to the notion of human-level AI seem to apply also to AGI.
12. J.B. Pollack, "Mindless Intelligence", IEEE Intelligent Systems, vol. 21, no. 3, 2006, pp. 50-56. All references to Pollack are to this paper. (Pubitemid 44144102)
13. P.J. Hayes and K.M Ford, "Turing Test Considered Harmful", IJCAI-95 Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, Montreal, Quebec, August 20-25, vol. 1, Morgan Kaufman, 1995, p. 974.
14. B. Whitby, "The Turing Test: AI-s Biggest Blind Alley?", in P. Millican and A. Clark eds, The Legacy of Alan Turing, Vol. I, Machines and Thought, Oxford Univ. Press, 1996, p. 58.
15. R.M. French, "Subcognition and the Limits of the Turing Test", in P. Millican and A. Clark eds, 1996, p. 12.
16. D. Michie, "Turing-s Test and Conscious Thought", in P. Millican and A. Clark eds, 1996, p. 36.
17. P.R. Cohen, "If not Turing-s test, then what?", AI Magazine, vol. 26, no. 4 (April), 2005, p. 62.
18. N. Block, "The Mind as the Software of the Brain", in E.E. Smith and D.N. Osherson eds, An Invitation to Cognitive Science, Vol. 3, Thinking, 2nd edition, MIT Press, 1995, p. 378.
19. P.J. Hayes and K.M. Ford, 1995, pp. 972, 976.
20. K.M. Ford and P.J. Hayes, "On Computational Wings: Rethinking the Goals of Artificial Intelligence", Scientific American Presents, vol. 9, no. 4, 1998, p. 79.
21. A. Sloman, "The Mythical Turing Test", 26 May 2010, http://www.cs.bham.ac.uk/research/projects/cogaff/misc/turing-test.html. Whitby makes the same claim, saying -Turing himself was always careful to refer to "the game". The suggestion that it might be some sort of test involves an important extension of Turing-s claims- ("The Turing Test: AI-s Biggest Blind Alley?", p. 54). Likewise Narayanan states that -Turing did not originally intend his imitation game to be a test- ("The Intentional Stance and the Imitation Game", in P. Millican and A. Clark eds, 1996, p. 66). For criticism of these claims (and other objections to Turing in this and the companion volume).
22. see B.J. Copeland and D. Proudfoot, -The Legacy of Alan Turing-, Mind, vol. 108, 1998.
23. A. Sloman, "The Mythical Turing Test". Sloman says, -[F]ar from proposing a test to answer the question whether machines can think or whether machines are intelligent, [Turing] actually decides (rightly) that the question is absurd- (ibid.). However, what Turing states is -absurd- is any attempt to answer this question by means of -a statistical survey such as a Gallup poll- ("Computing Machinery and Intelligence", Mind, vol. 59, 1950, p. 433). Despite his notorious remark that the question -Can machines think?- is -too meaningless to deserve discussion- (ibid., p. 442), Turing discussed this question at length (Copeland points this out in B.J. Copeland ed., The Essential Turing, Oxford Univ. Press, 2004, pp. 476-477). He discussed it in his 1950 paper (where, after all, he said that the question -Are there imaginable digital computers which would do well in the imitation game?- is a -variant- of the question -Can machines think?- (p. 442)), and in his radio broadcasts -Can Digital Computers Think?- (A.M. Turing, 1951, "Can Digital Computers Think?", in B.J. Copeland ed., 2004) and "Can Automatic Calculating Machines Be Said To Think?" (A.M. Turing, R. Braithwaite, G. Jefferson, and M. Newman, 1952, in B.J. Copeland ed., 2004). In the first of these broadcasts, for example, Turing speaks of -programming a machine to think- and -the attempt to make a thinking machine-, and he makes it clear that he is in favour of -the theory that machines could be made to think- (pp. 485-486).
24. A.M. Turing, 1950, pp. 452, 436, 446.
25. A.M. Turing, et al., 1952, p. 495.
26. A.M. Turing, 1948, "Intelligent Machinery", in B.J. Copeland ed., 2004, p. 431.
27. For my analysis of Turing-s approach to intelligence, see D. Proudfoot, "Rethinking Turing-s test", forthcoming.
28. D. Proudfoot and B.J. Copeland, "Artificial Intelligence", in E. Margolis, R. Samuels, and S. Stich eds, Oxford Hand-book of Philosophy and Cognitive Science, Oxford Univ. Press, forthcoming.
29. B.J. Copeland and D. Proudfoot, "Artificial Intelligence: History, Foundations, and Philosophical Issues", in P. Thagard ed., Handbook of the Philosophy of Psychology and Cognitive Science, Elsevier, 2006.
30. On anthropomorphism in AI, see e.g. D. Proudfoot, "How Human Can They Get?", Science, vol. 248, 1999, p. 745.
31. "The Implications of an Externalist Theory of Rule-Following Behaviour for Robot Cognition", Minds and Machines, vol. 14, 2004, pp. 295, 302.
32. B.J. Copeland and D. Proudfoot, 2006, pp. 445-446.
33. A.M. Turing, 1950, pp. 456-457; "Intelligent Machinery: a Heretical Theory", c. 1951, in B.J. Copeland ed., 2004, p. 473.
34. A.M. Turing, 1948, pp. 424-429.
35. A.M. Turing, 1948, p. 412.
36. V. Braitenberg, Vehicles: Experiments in Synthetic Psychology, MIT Press, 1984, pp. 83, 68, 19, 46, 83, 46, 41.
37. D.W. Hogg, F. Martin, and M. Resnick, "Braitenberg Creatures", Epistemology and Learning Group Memo No. 13, MIT Media Laboratory, 1991, pp. 1-8.
38. M. Yamamoto, " -SOZZY-: A Hormone-Driven Autonomous Vacuum Cleaner", Proceedings of the International Society for Optical Engineering, vol. 2058, 1993, pp. 212-213.
39. D.C. Dennett, "When HAL Kills, Who-s to Blame? Computer Ethics", in D.G. Stork ed., Hal-s Legacy: 2001-s Computer as Dream and Reality, MIT Press, 1997, p. 358.
40. "The Practical Requirements for Making a Conscious Robot", Philosophical Transactions of the Royal Society of London, vol. 349, 1994, pp. 140, 141.
41. See, e.g., C. Breazeal, "Regulating Human-Robot Interaction using -emotions-, -drives- and facial expressions", 1998; http://citeseerx.ist. psu.edu/viewdoc/summary?doi=10.1.1.56.6651.
42. See, e.g., C. Breazeal and J. Velásquez, "Toward Teaching a Robot -Infant- using Emotive Communication Acts", in Proceedings of 1998 Simulation of Adaptive Behavior, workshop on Socially Situated Intelligence; http://www.ai.mit.edu/projects/sociable/publications.html.
43. L. Wittgenstein, Philosophical Investigations, 2nd edition, eds. G.E.M. Anscombe and R. Rhees, Blackwell, 1972.
44. C. Breazeal, "Affective Interaction between Humans and Robots", in J. Kelemen and P. Sosik eds, ECAL 2001, LNAI 2159, Springer-Verlag, 2001, p. 585.
45. C. Breazeal and B. Scassellati, "Challenges in Building Robots That Imitate People", in K. Dautenhahn and C. Nehaniv eds, Imitation in Animals and Artifacts, MIT Press, 2001, www.ai.mit.edu/projects/humanoid-robotics-group/ publications.html.
46. C. Breazeal, 2001, p. 584. Not all human smiles and frowns are genuine, but non-genuine (-Duchenne-) smiles nevertheless have meaning, and the person smiling has a certain communicative intent.
47. C. Breazeal and P. Fitzpatrick, "That Certain Look: social amplification of animate vision", Proceedings of the AAAI Fall Symposium, Socially Intelligent Agents-The Human in the Loop, 2000, http://people.csail. mit.edu/paulfitz/publications.html.
48. D. McDermott, "Artificial Intelligence Meets Natural Stupidity", SIGART Newsletter, no. 57, April 1976, p. 4. Sloman makes the similar claim that referring to hypothesized parts of brains as -perceiving- or -deciding- is circular (A. Sloman, "The Design-Based Approach to the Study of Mind (in humans, other animals, and machines), Including the Study of Behaviour Involving Mental Processes", http://www.cs.bham.ac.uk/research/ projects/cogaff/misc/designbased-approach.html, February 28, 2010). I am indebted to Aaron Sloman for emphasizing the relevance of McDermott-s paper to the discussion of anthropomorphism in AI.
49. D. McDermott, 1976, pp. 4, 5.
50. Another example of plainly false ascriptions in AI is the use of personal pronouns (-he- and -she-) to refer to machines. A recent case is E.S. Brunette, R.C. Flemmer, and C.L. Flemmer, "A review of artificial intelligence", ICARA 2009 - Proceedings of the Fourth International Conference on Autonomous Robots and Agents, 2009, IEEE. S.P. 0Franklin goes so far as to use personal pronouns to refer, not only to robots and software agents, but also to processors and computer models (in Artificial Minds, Bradford Books, 1995).
51. Disease- is McDermott-s term (1976, p. 5).
52. D. McDermott, 1976, p. 4.
53. A.M. Turing, 1948, p. 412.
54. A.M. Turing, 1948, p. 431.
55. The judges- inputs quoted here are from the transcripts of, respectively, the 2005, 2003, and 2008 competitions; http://www.loebner.net/Prizef/ loebnerprize.html.
56. B.J. Copeland, "The Turing Test", in J.H. Moor ed., The Turing Test: The Elusive Standard of Artificial Intelligence, Kluwer, 2003, p. 7.
57. J.H. Moor, "The Status and Future of the Turing Test", in J.H. Moor ed., 2003, p. 204. The results of the 2000 contest are set out in "The Status and Future of the Turing Test", p. 205.
58. http://www.loebner.net/Prizef/loebner-prize.html.
59. Might an interrogator-s tendency to anthropomorphize somehow be triggered only (or more strongly) in her interviews with the machine? This possibility points to the fact that the imitation game, however interpreted, cannot be used as a one-off test. Individual interrogators may have quirks and may make surprising decisions; to obtain a convincing result, the imitation game must be played several times (see B.J. Copeland ed., 2004, pp. 528-529). There is no reason to think that, over a series of games using different interrogators chosen randomly, the tendency to anthropomorphize will favour the machine against the human contestant.
60. K.M. Ford and P.J. Hayes, 1998, p. 79.
61. N. Block, "Psychologism and Behaviorism", Philosophical Review, vol. 90, no. 1, 1981, p. 10.
62. See, e.g., M. Minsky, P. Singh, and A. Sloman, 2004.
63. A.M. Turing, 1950, p. 460.
64. For example, Turing-just as Pollack advocates-sought to understand how the human brain might emerge naturally; he said that -[t]he brain structure has to be one which can be achieved by the genetic embryological mechanism- and hoped that his work on neuron-like computation might clarify this process. (Letter from Turing to the biologist J.Z. Young. A copy is in the Modern Archive Centre, King-s College, Cambridge (catalogue reference K1.78)
65. B.J. Copeland ed., 2004, ch. 6 and pp. 353-355
66. Copeland B.J. Proudfoot D., 2006 On Neural Networks, see Turing, 1948.
67. reprinted in R. Chrisley ed., Artificial Intelligence: Critical Concepts in Cognitive Science, Volume 2: Symbolic AI, Routledge, 2000
68. For example, in addition to his ground-breaking work on mechanized search in the Bombe, Turing anticipated computation by neural networks and was the first to use computer simulation to investigate the development of pattern in living things. On the Bombe, see B.J. Copeland ed., 2004, ch. 6 and pp. 353-355; B.J. Copeland and D. Proudfoot, 2006. On neural networks, see Turing, 1948; B.J. Copeland and D. Proudfoot, "On Alan Turing-s Anticipation of Connectionism", Synthese, vol. 108, 1996 (reprinted in R. Chrisley ed., Artificial Intelligence: Critical Concepts in Cognitive Science, Volume 2: Symbolic AI, Routledge, 2000); B.J. Copeland and D. Proudfoot, "Alan Turing-s Forgotten Ideas in Computer Science", Scientific American, vol. 280 (April), 1999.
69. On simulation, see A.M. Turing, "The Chemical Basis of Morphogenesis", Philosophical Transactions of the Royal Society of London, Series B, vol. 237, 1952 (reprinted in B.J. Copeland ed., 2004; see also pp. 507-518);
70. B.J. Copeland and D. Proudfoot, "Turing and the computer", in B.J. Copeland ed., Alan Turing-s Automatic Computing Engine, Oxford Univ. Press, 2005.
71. K.M. Ford and P.J. Hayes, 1998, p. 79.
72. A.M. Turing, 1950, p. 435.
73. Discussions of philosophical objections to the Turing test include: J.R. Searle, "Minds, Brains, and Programs", Behavioral and Brain Sciences, vol. 3, 1980.
74. R.M. French, "The Turing Test: The First 50 Years", Trends in Cognitive Science, vol. 4, no. 3, 2000.
75. D. Lenat, "Building a Machine Smart Enough to Pass the Turing Test: Could We, Should We, Will We?"
76. P.M. Churchland, "On the Nature of Intelligence: Turing, Church, von Neumann, and the Brain", in R. Epstein, G. Roberts, and G. Beber eds, 2008.
77. D. Proudfoot, "Wittgenstein-s Anticipation of the Chinese Room", in J. Preston and M. Bishop eds, 2002 and "Review of James Moor (Ed.), The Turing Test: The Elusive Standard of Artificial Intelligence", Philosophical Psychology, vol. 19, no. 2, 2005.
78. B.J. Copeland and D. Proudfoot, 2006 and "The Turing Test: A Philosophical and Historical Guide," in R. Epstein, G. Roberts, and G. Beber eds, 2008.
79. See, e.g., N. Block, "The Mind as the Software of the Brain", p. 379. Block himself points out the discrepancy between the early Loebner contests and Turing-s test, yet describes Loebner-s test as a -reasonable facsimile- of Turing-s test.
80. N. Block, "The Mind as the Software of the Brain", p. 383.
81. The objection in this paragraph is based on B.J. Copeland, 2003, pp. 14-15.
82. N. Block, "The Mind as the Software of the Brain", p. 381.
83. On this point, see e.g. J.H. Moor, "Turing Test", in S.C. Shapiro ed., Encyclopedia of Artificial Intelligence, John Wiley, 1987, p. 1126.
84. A.M. Turing, et al., 1952, p. 494; B.J. Copeland first cites this quotation, to underline the flaw in the orthodox interpretation of Turing-s test, in his 2003, p. 6. Many critics misconstrue the test by treating it as an operational definition; a recent example is in Stevan Harnad-s running commentary on "Computing Machinery and Intelligence", in R. Epstein, G. Roberts, and G. Beber eds, 2008, p. 378.
85. A.M. Turing, et al., 1952, pp. 503-504. See B.J. Copeland, 2003, p. 15.
86. The recent philosophical literature on consciousness, zombie thought experiments, and machine consciousness is extensive. See, for example, D. Chalmers, "Facing up to the problem of consciousness", Journal of Consciousness Studies, vol. 2, no. 3, 1995.
87. N. Block, "On a confusion about the function of consciousness", Behavioral and Brain Sciences, vol. 18, no. 2, 1995.
88. J. McCarthy, "Todd Moody-s zombies", Journal of Consciousness Studies, vol. 2, no. 4, 1995;
89. D. Dennett, -The unimagined preposterousness of zombies-, Journal of Consciousness Studies, vol. 2, no. 4, 1995.
90. O. Flanagan and T. Polger, "Zombies and the function of consciousness-, Journal of Consciousness Studies, vol. 2, no. 4, 1995.
91. K. Balog, "Conceivability, Possibility, and the Mind-Body Problem", Philosophical Review, vol. 108, no. 4, 1999.
92. R. Kirk, "The inconceivability of zombies", Philosophical Studies, vol. 139, 2008.
93. A. Sloman, "Phenomenal and Access Consciousness and the -Hard- Problem: A View from the Designer Stance-, International Journal of Machine Consciousness, vol. 2, no. 1, 2010.
94. J.R. Searle, "Twenty-One Years in the Chinese Room", in J. Preston and M. Bishop eds, Views into the Chinese Room: New Essays on Searle and Artificial Intelligence, Oxford Univ. Press, 2002, pp. 52, 64.
95. -Intelligence amplifier- is Lenat-s term. D.B. Lenat, "Building a Machine Smart Enough to Pass the Turing Test: Could We, Should We, Will We?", in R. Epstein, G. Roberts, and G. Beber eds, Parsing the Turing Test: Philosophical and Methodological Issues in the Quest for the Thinking Computer, Springer, 2008, p. 281.
96. See, for example, P.R. Cohen, "If not Turing-s test, then what?".
97. A.M. Turing, et al., 1952, p. 495. This prediction has generally been ignored in the literature on the Turing test, in favour of Turing-s prediction in his 1950 paper, where he states that -in about fifty years- time it will be possible to programme computers, with a storage capacity of about 109, to make them play the imitation game so well that an average interrogator will not have more than 70 per cent. chance of making the right identification after five minutes of questioning- (A.M. Turing, 1950, p. 442). Emphasis on the much weaker 1950 prediction tends to distort the nature of Turing-s claims concerning his test.
98. See, for example, P. Hayes and K. Ford, 1995; J.B. Pollack, 2006; D.B. Lenat, "The voice of the turtle: whatever happened to AI?".
99. For example, A. Sloman, "The Mythical Turing Test".
100. A.M. Turing, c. 1951, p. 484; 1950, pp. 434, 435.
101. A. Sloman, "The Mythical Turing Test".
102. A. Sloman, "The well-designed young mathematician", Artificial Intelligence, vol. 172, 2008; "The Design-Based Approach to the Study of Mind (in humans, other animals, and machines), Including the Study of Behaviour Involving Mental Processes"; "An Alternative to Working on Machine Consciousness", http://www.cs.bham.ac.uk/research/projects/cogaff/09. html#910, May 14, 2010; and "Requirements for Artificial Companions: It-s harder than you think", http://www.cs.bham.ac.uk/research/projects/cogaff/ 07.html#711, April 25, 2010.
103. A.M. Turing, et al., 1952, p. 500.
104. R.A. Brooks, "Intelligence without Reason", in L. Steels and R.A. Brooks eds, The Artificial Life Route to Artificial Intelligence, Lawrence Erlbaum, 1995, p. 57.
105. R.A. Brooks, Cambrian Intelligence: The Early History of the New AI, Bradford Books, 1999, p. x.
106. D.C. Dennett, "Intentional Systems", in his Brainstorms: Philosophical Essays on Mind and Psychology, Harvester Press, 1979, pp. 3-4.
107. D.C. Dennett, 1979, p. 8.
108. Personal communication from Rodney Brooks.
109. A classic early text is S.E. Guthrie, Faces in the Clouds: A New Theory of Religion, Oxford Univ. Press, 1993.
110. D.L. Waltz, "Evolution, Sociobiology, and the Future of Artificial Intelligence", IEEE Intelligent Systems, vol. 21, no. 3, 2006, p. 68.