Editor’s note: This interview, although based on a set of quotes reportedly by Alan Turing, is entirely fictitious.
Is it enough that a computer can be called intelligent if it can deceive a human into believing that the computer is human? Maybe not. The film Ex Machina conceptually extended the Turing Test of artificial intelligence (AI)—as proposed by Alan Turing during the mid-20th century—to the point where the question becomes not only about whether AI can deceive a human, but also whether AI can determine human actions, such as falling in love.
The subject of AI and whether AI really is intelligent is complex. But what exactly is AI? Let us ask the veritable father of computer science, Alan Turing, some questions to learn more.
Guy: “Welcome, Alan, and thank you for bringing your experience and insights into today’s conversation about AI. For a question to warm things up, could AI be considered as something having a computer that replicates the thought processes of an adult human brain?”
Alan Turing: “[Well,] [i]nstead of trying to produce a program to simulate the adult mind, why not rather try to produce one which simulates the child’s [brain]? If this were then subjected to an appropriate course of education, one would obtain the adult brain.”
Guy: “That is interesting, Alan. That would mean that AI is influenceable and malleable as a subject of various environmental stimuli and that AI is always learning and adjusting its perspective according to these stimuli. A test of whether something is AI could thus also test the extent to which the AI learns and unlearns things. On that point, can AI always be expected to be right given the dynamics of the process of learning and unlearning?”
Alan Turing: “[I think that] [i]f a machine is expected to be infallible, it cannot also be intelligent.”
Guy: “So AI that does not make mistakes may not be deemed intelligent, at least with human intelligence as a frame of reference. In other words, the vagaries of learning and unlearning is an even more critical trait of intelligence than we may think, but what abo…”
Alan Turing: “[Hold on, Guy, I would like to add that] [u]nless in communicating with it one says exactly what one means, trouble is bound to result.”
Guy: “But where is the intelligence in that? If AI cannot manage this kind of ambiguity as the human brain can, then just how intelligent is AI really?”
Alan Turing: “[Interesting question, Guy. Let us change the subject.] It seems probable that once the machine thinking method had started, it would not take long to outstrip our feeble powers … They would be able to converse with each other to sharpen their wits. At some stage therefore, we should have to expect the machines to take control.”
Guy: “Hmm, that scenario sounds reminiscent of a film beyond your time horizon on Earth, 2001: A Space Odyssey, where the spaceship’s AI computer, HAL, took over. A turning point in that movie emerged when HAL beat astronaut Dave Bowman at a game of chess. Interestingly, Alan, you created the first algorithm for computer chess before there was even a computer to use it on. It is interesting how often conversations about chess and AI seem to cross paths. Indeed, such an event will happen about a decade beyond your life’s time horizon between IBM’s Deep Blue and Garry Kasparov, and another between the same opponents a further decade later.”
Alan: “[So I will be long gone when that all happens? That sounds amazing! You guys get to have all the fun.]”
Guy: “That may be true! Thank you for sharing your extraordinary and visionary insights with us, Alan. Perhaps we will see your face on a £50 note or a day in commemoration of you!”
Alan: “[Well, that would be a change in government’s perception of me! I understand the government will apologize in 2009, and Queen Elizabeth will eventually pardon me in 2013. All a bit late, but perhaps rather late than never.]”
Ultimately, AI is a malleable portfolio of algorithms that themselves are malleable. But if algorithms are so volatile, that poses a host of challenges not only for the board, but also for management.
Editor’s note: For further insights on this topic, read Guy Pearce and Maureen Kotopski’s recent Journal article, “Algorithms and the Enterprise Governance of AI”, ISACA Journal, volume 4, 2021. Learn more about AI fundamentals as part of ISACA’s Certified in Emerging Technology (CET) credential.
Don't forget—Members can earn free CPE from ISACA Journal quizzes!