Yzabel / July 17, 2017

Review: Turing’s Imitation Game

Turing's Imitation Game: Conversations with the UnknownTuring’s Imitation Game: Conversations with the Unknown by Kevin Warwick

My rating: [rating=3]

Blurb:

Can you tell the difference between talking to a human and talking to a machine? Or, is it possible to create a machine which is able to converse like a human? In fact, what is it that even makes us human? Turing’s Imitation Game, commonly known as the Turing Test, is fundamental to the science of artificial intelligence. Involving an interrogator conversing with hidden identities, both human and machine, the test strikes at the heart of any questions about the capacity of machines to behave as humans. While this subject area has shifted dramatically in the last few years, this book offers an up-to-date assessment of Turing’s Imitation Game, its history, context and implications, all illustrated with practical Turing tests. The contemporary relevance of this topic and the strong emphasis on example transcripts makes this book an ideal companion for undergraduate courses in artificial intelligence, engineering or computer science.

Review:

[I received a copy of this book through NetGalley.]

That was an informative, albeit also controversial, read about Turing’s ‘Imitation Game’, focused on the game itself rather than on the man (who I like reading about in general, but here I was definitely more interested in his famous ‘test’, since I keep hearing about it, but never in much detail). It sheds light on Turing’s aim when devising the test, as well as on what he predicted, and that may or may not happen sooner than expected.

Several sections in the book are devoted to examples of studies and events during which the test took place, pitching human judges against both machines and other human beings, without the former knowing what or who the latter was. Actual, textual examples allow the reader to try and make their own judgment—and determining where the machines are is not so easy as it seems. I was accurate in my guesses except but once, I think, however I can see where judges were ‘fooled’, and why. At other times, I was surprised at the outcome, for instance quite a few human participants made ‘boring’ answers to conversations, which in turn prompted judges to believe they were talking to a machine—and conversely, some AIs were clearly programmed with a variety of lively potential responses. Eugene Goostman, especially, with its persona of a 13-year old Ukrainian boy whose English is only second language, has good potential (in that you can tell some of its/his answers are stilted, but not more than if it/he was an actual learner of ESOL).

The test as a whole posits several interesting questions and conundrums. Namely, the fact that it’s based on language, and that one may wonder whether being able to converse means one is gifted with ‘thought’. Another one is whether the test as it exists can really be used as a marker: aren’t the various chatbots/AIs out there simply well-programmed, but in no way indicative of whether they’ll be able to go further than that?

Also, I’m not sure I can agree with the 2014 ‘the Turing test has been passed’ result, as it seems to me the percentage is too low to warrant such a qualifier (if 90% of judges were fooled in believing they were conversing with a human, now that’d be something else… or am I aiming too high?), and it’s too early anyway for the current AIs to have been developed far enough (as fascinating as some of their conversations were, they still looked much more like complex chatbots than anything else—at least, to me).

Conclusion: 3.5 stars. I did learn quite a few things no matter what.