Kkumar@wcupa.edu
A review of Kissinger, H. A., Schmidt, E., & Huttenlocher, D. (With Schouten, S., [2021]). The Age of AI and our human future. Little Brown and Company. 254 pp., $30
When I saw this book at Barnes and Noble on sale for a 50% discounted price, I couldn’t resist buying it because it was written by Henry Kissinger (who turned 100 on May 27, 2023), our 56th Secretary of State, and two prominent computer scientists: Eric Schmidt, former Google CEO, technological entrepreneur, philanthropist, and executive chairman and advisor at Google, and Daniel Huttenlocher, the inaugural dean of the MIT Schwarzman College of Computing.
Written for people who know little about AI, there is much to learn from this book: definitions, basic terminology, capabilities, future capabilities, philosophical questions, the notion of machine intelligence, how a machine learns, and broader socio-cultural-geopolitical (including national security) implications and concerns that arise from the implementation of rapidly advancing AI technologies. It has 7 chapters: Where We Are; How We Got There, Technology and Human Thought; From Turing to Today and Beyond; Global Network Platforms; Security and World Order; AI and Human Identity; and AI and the Future.
In a book review of James Gleick’s (2011) The Information: A History, A Theory, A Flood, I noted that Gleick described Charles Babbage’s “remarkable wish” expressed “some years before his death [1871], he told a friend that he would gladly give whatever time he had left if only he could be allowed to live three days, five centuries later” (Gleick, p. 124). And I concluded my review of Gleick’s book by stating, “I would settle for three days a mere 50 years in the future!” (Kumar, 2012). I feel lucky to have been able to live in the age of AI.
To paraphrase Charles Dickens, reading this book makes it clear that we are living in the most exciting of times, and perhaps the most worrisome of times, of technological development where what was the stuff of science fiction years is fast becoming an everyday reality. Who would have thought 20 years ago that the ubiquitous “all-in-one” encyclopedic cellphone, with apps powered by AI, would become the bread and butter of everyday communications, information, misinformation, disinformation, navigation, ridesharing, shooting and editing photos and videos, entertainment, banking, and much more? Kissinger et al. stress, AI is everywhere (e.g., Alexa, Siri, Google Assistant, social media, automated vehicles, tools, and appliances) making our everyday tasks putatively easier and safer. The OpenAI ChatGPT, widely talked about in the news lately, uses a conversational format to answer questions, write your school essay, translate a passage from another language, show the excel code to compute correlations, etc. The OpenAI ChatGPT possibly could have written this review or edited it (I didn’t try either asking to write it or edit it, but I cut and pasted two lines from a Hindi poem [in Hindi], and it came up with a reasonable translation.)
Kissinger et al. claim, “This book is about a class of technology that augurs a revolution in human affairs—machines that can perform tasks that require human-level intelligence—has rapidly become a reality” (p. 14). A major consequence of AI-enabled technology, they note, is that we have already begun to see the world in a different way—it has altered almost every aspect of our lives, including the way we think and do everyday things. We have begun to rely on machine logic to inform us about what is happening in the world, public opinion, what to read, gathering information for writing, where and what to shop, what movies to see, what music to listen to, what stocks to buy and sell, where to travel and what to see, getting psychological therapy, etc. AI has mastered chess, discovered new drugs, developed writing programs, and employed in gathering intelligence and conducting war.
Even though humans have developed AI, it now controls many aspects of our life—perhaps even without our realizing it. Per Kissinger et al., a decade ago, search engines employed data mining, now they are powered by machine learning. The AI-enabled search engines learn what you and similar others searched for online, bought, read, etc. and use that information to direct you (or steer you away from) to certain products/brands/content. According to Kissinger et al., if you searched for gourmet food, you might be directed to designer clothing when you search for clothing. The footprints you leave on the Internet become the basis of AI predictions. It is no wonder you can find articles titled “Why the internet knows us better than We Know Ourselves” or “The Internet knows You Better Than Your Spouse Does.”
Kissinger et al. note “AI promises to transform all realms of human experience. And … its transformation will ultimately occur at the philosophical level, transforming how humans understand reality and our role within it” (p. 17). They stress that advances in AI combined with its increased availability are likely to revolutionize and redefine science, medicine, communications, transportation, art, culture, geopolitics, warfare, and deterrence strategies. An interesting question is as our dependence on AI-enabled decisions (e.g., reliance on GPS, self-driving cars) increases, would it lead to changes in our cognitive capacities in navigating our environments? Have we already lost our ability to read maps? Now we learn how to download and operate an app such as Waze.
The book discusses AI’s terminology issues, some of its current contributions, potential uses, consequences, limitations, and human limitations in designing it and making sense of its results. Major concerns with AI include our personal security, the security of our borders, and its possible deployment in cyber warfare, AI-enabled warfare, and nuclear warfare. Who will control AI warfare technologies? Can we prevent rogue individuals and countries from developing such technologies or using popular social media apps to conduct hybrid warfare on a large scale? The old concept of arms control by nations takes on a new meaning when it comes to AI-enabled warfare, particularly when employing automated decision-making by AI. The book discusses many of these matters in detail.
Kissinger et al. highlight that if AI is given autonomous warfare decision-making capabilities, we would need new international treaties and laws to regulate weaponized AI because this capability would challenge the traditional notions of “defense and deterrence” (p. 21). They indicate AI users will not be just large corporations, or governments, but also individuals. While large corporations and governments tend to act by regulations or with restraint, it is difficult to predict how individuals will use AI-enabled software to benefit society or to disrupt everyday life, or in an extreme scenario some rogue individual or state may try to take over the world in the manner of Dr. Evil, in Mike Myers’ Austin Powers series!
It is difficult to predict where this technology will lead us. But it will give us a lot to think about and work with. As the book’s authors stress that once trained, AI works using its own logic “that is nonhuman and, in many ways, inscrutable to humans” (p. 107). Thus, they note AI can explore “aspects of reality we have never known and may never directly know” (p. 16), and its results may not always be consistent, straightforward, or meaningful, but they will allow us to ask questions. Therefore, the authors stress that we need to partner with AI to make it work for us. Currently, there is much discussion on students using ChatGPT to cheat (see Christopher Grobe, 2023). Soon, however, AI may write a book or a scientific abstract (Elise, 2023) that the reader (or a machine algorithm) may not be able to decipher if it was written by a human or an intelligent machine—perhaps meeting the criterion for Alan Turing’s test, not so much in terms of performance indistinguishability from humans, but “whose performance is humanlike” (p. 56; italics in original).
The book is timely and worth every penny (of its original price, not the discounted price I paid)! It is neither a long book nor an overly technical book and there is much to learn about AI from this book.
References
Gleick, J. (2011). The Information: A History, A Theory, A Flood. Pantheon Books.
Kumar, V. K. (March 28, 2012). From drumbeats to bits to cubits: The dawn of the e-age. (A review of Gleick’s book.) PsycCritiques, 57(13), Article 6.
