7 Classic Books To Deepen Your Understanding of (Artificial) Intelligence

M.C. Escher's artwork was an inspiration for Douglas Hofstadter's 1979 book "Gödel, Escher, Bach: An ... [+]

M.C. Escher

The field of artificial intelligence has never been the subject of more attention and analysis than it is today. Almost every week, it seems, a new bestselling book comes out examining the technology, business or ethics of AI.

Yet few of the topics and debates at the center of today's AI discourse are new. While not always recognized by commentators, artificial intelligence as a serious academic discipline dates back to the 1950s. For well over half a century, many of the world's leading minds have devoted themselves to the pursuit of machine intelligence and have grappled with what it would mean to succeed in that pursuit.

Much of the public discourse around AI in 2019 has been anticipated—and influenced—by AI thought leaders going back decades.

Below is a selection of seven classic books about intelligence: what it is, how we might build machines that have it, and what that would mean for society. These books have played a formative role in the development of the field of AI; their influence continues to be felt today. For anyone seeking a deep understanding of AI's complexities, challenges, and possibilities, they are essential reading.

Gödel, Escher, Bach: An Eternal Golden Braid (Douglas Hofstadter, 1979)

Gödel, Escher, Bach is sometimes referred to as “the Bible of artificial intelligence” (though Hofstadter himself rejects the label).

 

The book's central theme is that, through self-reference and “strange loops”, systems comprised of independently meaningless elements can acquire meaning and intelligence. Hofstader identifies versions of such recursive systems in fields as diverse as mathematics, music, art and computer science.

To sketch out his subtle thesis, Hofstadter takes his reader into the depths of number theory, classical music and the computing technology stack; he employs fanciful dialogues between fictional characters in the style of Lewis Carroll; he structures the book’s chapters, paragraphs and sentences to themselves embody his points about recursion.

Though Hofstadter was an unknown author at the time, Gödel, Escher, Bach won both the Pulitzer Prize and the National Book Award.

The Singularity is Near (Ray Kurzweil, 2005)

Perhaps no book or author presents a more relentlessly optimistic view of our technological future than Ray Kurzweil's The Singularity Is Near.

Grounding his arguments in the concepts of exponential growth and accelerating returns, Kurzweil anticipates a future in which machine intelligence compounds and soon far outstrips today’s human-level intelligence. He foresees human intelligence transcending biology, adapting to non-biological substrates and eventually spreading throughout the universe.

While his conclusions are startling, his approach is meticulously data-driven, with extensive historical analysis of the growth of computational capabilities over time. As the New York Times put it, “Kurzweil's vision of our super-enhanced future is completely sane and calmly reasoned.”

The concept of the Singularity has inspired generations of technologists; it has also garnered plenty of ridicule for its fantastical, utopian overtones. Kurzweil did not invent the idea, but he and this book have played a major role in popularizing it.

Alan Turing: The Enigma (Andrew Hodges, 1983)

It is only a slight overstatement to say that Alan Turing created the computer and the field of artificial intelligence. His seminal 1936 paper, vastly ahead of its time, laid the conceptual groundwork for the entire field of digital computing. He was one of the first thinkers to take the idea of artificial intelligence seriously. His 1950 paper—which famously opens with the line “I propose to consider the question, ‘Can machines think?’”—introduced the Turing Test, which remains a touchstone in the AI literature today.

Andrew Hodges' 1983 work is the authoritative biography of Turing's life. Prior to its publication, Turing's accomplishments were not widely known, due largely to the total secrecy that for decades surrounded his wartime work on cryptography for the Allies at Bletchley Park. Hodges’ book played a pivotal role in bringing Turing’s ideas to light and establishing him at the forefront of the pantheon of machine intelligence pioneers.

On the topic of AI, Turing made it clear where he stood. Generations ahead of his time, in words that would still be provocative today, Turing wrote in 1951: “It is customary to offer a grain of comfort, in the form of a statement that some peculiarly human characteristic could never be imitated by a machine. I cannot offer any such comfort, for I believe that no such bounds can be set.”

Descartes' Error: Emotion, Reason, and the Human Brain (Antonio Damasio, 1994)

Conventional wisdom has long held that, while the intellect is logic-based and objective, emotions make us irrational and cloud our judgment.

In Descartes’ Error, neurologist Antonio Damasio famously reconceptualized the relationship between emotion and intellect. The book argues that emotions in fact play an essential role in cognition and decision-making, and that without them our intellectual capabilities would not be possible.

This theory of intelligence has intriguing implications for AI. As Marvin Minsky once put it: “The question is not whether intelligent machines can have any emotions, but whether machines can be intelligent without any emotions.”

On Intelligence (Jeff Hawkins, 2004)

In On Intelligence, Jeff Hawkins posits that a single fundamental “algorithm” underlies all information processing in the human brain: a feedforward mechanism that predicts future states.
Hawkins’ theory of intelligence has been highly influential across neuroscience, machine learning and philosophy over the past 15 years. It has also been regularly criticized. In 2005 Hawkins founded the AI startup Numenta with Dileep George to pursue these principles.

The Society of Mind (Marvin Minsky, 1986)

Marvin Minsky is one of the founding fathers of artificial intelligence. The Society of Mind, his most famous and readable work, lays out his perspectives on how the human mind works and how we might build machines that simulate it.

Minsky's overarching thesis is that intelligence emerges from the interactions of countless non-intelligent “agents” functioning in synchrony, generating outcomes more complex than the sum of its parts.

“What magical trick makes us intelligent?” he writes at the end of the book. “The trick is that there is no trick. The power of intelligence stems from our vast diversity, not from any single, perfect principle.”

The Mind's I (Douglas Hofstadter and Daniel Dennett, 1981)

In The Mind's I, Hofstadter and Dennett undertake to examine the most fundamental of questions: what is thought, what is consciousness and what is the mind? They do so through an annotated anthology of pieces from contributors as diverse as Richard Dawkins, Jorge Luis Borges and Alan Turing.

The book contains rich insights about what it would mean for a machine to think, interweaving perspectives from psychology, engineering, philosophy and literature. But don’t expect to walk away with any straightforward answers. As they write in the preface: “We believe there are at present no easy answers to the big questions. This book, then, is designed to provoke, disturb, and befuddle its readers.”