AI is one of the hottest topics of our time, sometimes promising a sunlit future of leisure and at other times threatening destruction by an army of bots indistinguishable from humans. In fact - might they actually be human? This highly informative book busts some of the myths while challenging the reader. Starting with so-called chess-playing machines of the 19th century which actually concealed a human inside, the author moves on to world champion Kasparov's battles against IBM's Deep Blue, a genuine machine, which finally managed to beat Kasparov. AI panic is widespread but so far only 2 of the 270 jobs in the 1950 US census have been replaced mechanically. Robots can take over part of many jobs but not the whole thing. In 1950 Alan Turing asked the question, "Can machines think?" and developed the Turing test to identify whether responses come from a human. Disturbingly, though, getting things wrong is part of the way bots are now programmed. ChatGPT was an immediate internet success in 2022 and is one of the largest neural networks built to date. Many of the tasks it could do were unexpected even to the developers. Faking creativity is a key area, and AI can create a passable though not brilliant Eurovision entry and pictures of cats in the style of Picasso, Van Gogh and Rembrandt. The author believes that the capacity of AI to produce fakes and other forms of deception is limiting its more legitimate development to build machines that will take over dull, dangerous and difficult jobs. Turing's goal of building a computer that could pretend to be human was the wrong one. AI needs a reset, and the author suggests ways it could be done. 234pp, diagrams.
Additional product information
