Tom Chivers in his book The AI does not hate you: Superintelligence, rationality and the race to save the world deals with such scenarios.
For the Avengers fans, there is no more significant moment than Tony Stark (Iron man) and Dr Banner (Hulk) experimenting with next-level intelligence. In their two experiments, they end up with Ultron and Vision. Two sides of the same coin. What ensues may be a long-drawn fight, but for technologists, it has become a quintessential question. Will we end up with an Ultron or a Vision? It is one that Jack Ma and Elon Musk also dealt with in a recent debate. While Ma was more welcoming of AI, Musk fears that it may end up destroying humanity. While our fears are well-rooted in science fiction, take Terminator, or I-Robot, for instance, there is more to AI than meets the eyes. Logic would tell you as long as you are controlling the machine, it shouldn’t matter, and there are still some years to creating an apocalypse, but the issue needs more discussions, especially, when we are talking about creating or dealing with intelligence superior to ours.
Tom Chivers in his book The AI does not hate you: Superintelligence, rationality and the race to save the world deals with such scenarios. A journalist by profession, he does so by looking at and interviewing a specific class of people called the rationalists. Trying to decode how people are trying to decipher technology and intelligence from a different perspective, those creating technology may not entirely see that. The book describes the history of the movement and how it’s come to attract more people. But more often than not, it deals with the psychology that goes behind such initiatives and the creation of AI. It’s not a complete doomsday scenario, and the author does build humour well, right throughout his book. But that is not to say that Chivers shies away from revealing the frightening details of AI implosion and explosion.
However, the timeline becomes an essential feature of the book, which takes away a good part of the argument, and every discussion starts with a morbid view of the technology. While the author does provide exciting insights, sometimes the arguments seem to be going nowhere, before you can finally make sense of it all. It’s like a thriller, only that you know what the conclusion is going to be, i.e., if there is one. My only gripe with books on technology is that they only detail one side of the argument, Chivers has done well to avoid that. While it’s true we are dealing with the unknown while contemplating about AI, and AI may not hate you, but it may not love you either.