The summer is already in full swing, and so are the summer holidays. Can you imagine a perfect summer holiday break without a book? We certainly can’t, and so we recommend five books for curious minds to indulge in this summer.
KLARA AND THE SUN by Kazuo Ishiguro
A fiction novel written by the Nobel Prize-winning author Kazuo Ishiguro explores the relationship between AI and human. The novel tells the story from the perspective of Klara, an Artificial Friend, who carefully observes the behavior of everyone around her in order to understand the complicated human relationships and such abstract terms as friendship, family bonds, and love. The story also touches upon interesting moral and scientific questions. For example, whether to “upgrade” children using genetic-engineering, which could allow them to adapt better to an advanced society, however, you can risk their health or even their life; also whether you can completely re-create your child if you replicate the same body and replace the mind of your child with AI. Then what is the difference between humans and robots/AI if you have the same appearance and AI can learn to copy our behavior, emotions, and speech? What makes us special and which one would be a real version of us? And here Klara gives an interesting suggestion of what could make us different from AI/robots:
“Mr Capaldi believed there was nothing special inside Josie that couldn‘t be continued. He told the Mother he‘d searched and searched and found nothing like that. But I believe now he was searching in the wrong place. There was something very special, but it wasn‘t inside Josie. It was inside those who loved her.”
If you are already curious or don‘t fully understand the quote, better read an entire story.
THE ROAD TO CONSCIOUS MACHINES: THE STORY OF AI by Michael Wooldridge
Michael Wooldridge, a Professor of Computer Science at the University of Oxford, discusses the idea that building an intelligent AI is a challenging task. What makes it so problematic? First, if set out to build intelligent machines, then intelligence must ultimately be reduced to simple explicit instructions, such as add A to B, repeatedly do F until G. Thus, the question of whether general AI is possible amounts to whether we can produce intelligent behavior by following lists of instructions like these. However, such simple logic often fails in real life because we encounter too many ambiguous situations, e.g., taxes are good and bad, wine may taste good and bad at the same time, etc. This becomes a problem when we are trying to use logic for something it was not intended for, because in math if you encounter a contradiction, it means that you made a mistake, however, this is not necessary the case in our lives. Similarly, we often fail to communicate our desires to a computer. First, we sometimes ourselves cannot explicitly know what we want. Second, even the experts in the field often find it difficult to articulate their expertise – the fact that they are good at something does not mean that they can tell you how they actually do it. Finally, when we communicate with other humans, we share common values and norms, therefore, many things are based on our general knowledge or assumptions about environment, culture, and society (there are some very good examples presented in the book). Sometimes your computer can do what you asked for, but the result may not be what you anticipated.
If you are curious about the history and the future of the general AI, and even if you have no prior knowledge of AI or computer science, this book can be a useful introduction to the topic.
THE FUTURE OF HUMANITY & THE FUTURE OF THE MIND by Michio Kaku
Michio Kaku is an American theoretical physicist at the City College of New York, a best-selling author, a futurist, and a communicator and popularizer of science. He has written several books about astrophysics, science and the future.
“The Future of the Mind” takes us on a tour around world across the top laboratories to explore the latest achievements in neuroscience and physics. The book also explores the future possibilities of how these technologies and advancements could change our daily lives in the near and far future. Could we invent a “smart pill” to enhance our cognitive abilities? Could we establish a “brain-net” to control our minds and connect them to advanced computers and robots? Could we push the limits of our consciousness and consciousness of AI? Could we use telekinesis and telepathy? And what about an alien mind?
“The Future of Humanity” is an extension of the predictions of our future. This book involves not only our mind, but also details how humanity can use the developments in robotics, nanotechnology, and biotechnology to enable us to create habitable cities on Mars or other nearby stars, how it could allow us to sail through space and perhaps even entirely transcend our physical bodies. In both of these books you will find the mind-blowing predictions and possibilities that, at first glance, could sound like a science-fiction, however, they are based on the developments of science and technology.
ATLAS OF AI: POWER, POLITICS AND THE PLANETARY COST OF ARTIFICIAL INTELLIGENCE
Kate Crawford, a Research Professor of Communication and Science and Technology Studies at the University of the Southern California, among other affiliations, manages to show the side of AI that is rarely considered. When we think about AI, some of us may picture AI as disembodied structures of computer code. Crawford argues that AI systems are neither artificial, nor intelligent. In fact, they are very physical, with a heavy reliance on natural resources and human labour. Consider the minicomputer you keep at your side almost at all times, your smartphone. The creation of smartphone devices requires using a variety of metals, some of which are rare earth metals, such as yttrium, terbium, lanthanum, etc. Another example of the physical nature of high-tech is the human labour required to pre-process and annotate the data that can then be used to train algorithms. Machine learning algorithms that work well are extremely data-hungry. Yet these algorithms should not be taken to stand for human intelligence because essentially what they are good at is detecting patterns, clustering and making predictions, which fall short of human cognition. Finally, there are social practices at play that shape the creation of AI. The engineers that design AI systems are mostly males that come from similar socio-economic backgrounds. This homogeneity influences the problems that are chosen to be addressed using AI.
All this sounds like a social critique of AI. It is for a reader to decide how much this is a scathing critique and how much it is a depiction of current reality of what AI systems are built on.