Artificial Intelligence

Evolution, Ethics and Public Policy

Saswat Sarangi, Pankaj Sharma

• Éditeur: Routledge India
• Année: 2018
• ISBN:
  9780429865404
• Édition: 1
• Sujet: Politics & International Relations
• Sous-sujet: Public Policy

What will the future be?

A dystopian landscape controlled by machines or a brave new world full of possibilities?

Perhaps the answer lies with Artificial Intelligence (AI)—a phenomenon much beyond technology that has, continues to, and will shape lives in ways we do not understand yet.

This book traces the evolution of AI in contemporary history. It analyses how AI is primarily being driven by "capital" as the only "factor of production" and its consequences for the global political economy. It further explores the dystopian prospect of mass unemployment by AI and takes up the ethical aspects of AI and its possible use in undermining natural and fundamental rights.

A tract for the times, this volume will be a major intervention in an area that is heavily debated but rarely understood. It will be essential reading for researchers and students of digital humanities, politics, economics, science and technology studies, physics, and computer science. It will also be key reading for policy makers, cyber experts and bureaucrats.

1

AI—The history and evolution

I believe that at the end of the century the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted.

—Alan Turing¹

Intelligence: There is still no agreement on its definition

The definition of intelligence is highly subjective and there is hardly a consensus over what intelligence really means. It can be defined as the capacity for logic, understanding, self-awareness, learning, emotional knowledge, planning, creativity and problem solving. It can be more generally described as the ability or inclination to perceive or deduce information, and to retain it as knowledge to be applied towards adaptive behaviors within an environment.²

Merriam-Webster defines intelligence as the ability to learn or understand things or to deal with new or difficult situations.³ Albert Einstein once said, “The true sign of intelligence is not knowledge but imagination.” Socrates said, “I know that I am intelligent, because I know that I know nothing.” For centuries, people have tried to define intelligence.⁴ The following passage on intelligence is from “Mainstream Science on Intelligence” (1994), an op-ed statement in the Wall Street Journal signed by 52 researchers:⁵

A very general mental capability that, among other things, involves the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly and learn from experience. It is not merely book learning, a narrow academic skill, or test-taking smarts. Rather, it reflects a broader and deeper capability for comprehending our surroundings—“catching on,” “making sense” of things, or “figuring out” what to do.

The American Psychological Association⁶ notes that intelligence refers to intellectual functioning. Intelligence quotients, or IQ tests, compare your performance with other people your age who take the same test. These tests don’t measure all kinds of intelligence, however. For example, such tests can’t identify differences in social intelligence, the expertise people bring to their interactions with others. There are also generational differences in the population as a whole. Better nutrition, more education and other factors have resulted in IQ improvements for each generation.⁷

Natural intelligence originates in biology. Simply speaking, natural intelligence is how animal or human brains function. But, the definition of intelligence which links it to the brain alone is a narrow one as nature also demonstrates non-neural control in plants and protozoa and distributed intelligence in colony species like ants, hyenas and humans.⁸ However, for all practical purposes, human intelligence is referred to as natural intelligence.

Natural intelligence vs. artificially created intelligence

Human intelligence or natural intelligence is created naturally and biologically. This is different from intelligence which is created by humans in machines using technology. It is expected that there will be differences between natural intelligence and artificially created intelligence.⁹ Naturally, AI has significant dominance in many tasks, especially when it comes to monotonous judgments. In contrast, biological neural networks also have superiority in some qualities:

1. Speed – Natural intelligence will be at a disadvantage versus an artificially created intelligence and this is similar to the advantage a computer has over the human brain in performing calculations. Natural intelligence will also have more “down time”.
2. Objectivity – Humans have inherent biases in their decision-making and in most cases these biases can be eliminated by using an artificially created intelligence. The artificially created intelligence can also be more accurate and precise.
3. Handling complex and different tasks – Machines are usually designed for a few tasks while humans can do better at handling more complex and different types of activities. Humans will also have an advantage in multitasking.
4. Complex movements – Even the most advanced robots can hardly compete in mobility with humans. This makes humans more suitable for tasks that require a higher degree of maneuverability.
5. Evolution – The pace of evolution in humans is very slow and it takes thousands of years to make the changes and modifications required as per the change in the environment. But machines can adapt very quickly.
6. Cost –Natural intelligence resides in humans and the life cycle cost of humans has not changed much over thousands of years when we talk about covering the basic necessities. However, the cost of creating, operating and maintaining machines is reducing at a fast pace.

The beginning of AI

Ideas and fantasies about machines that can think like humans have been around for a long time. One of the primary drivers of this thought process was the desire to create something that had all the human abilities but none of the human shortcomings. The idea was to create a machine which can think and act like a human even in subjective and ambiguous situations, but which does not get interrupted by mundane distractions and the machine also always remains consistent, unlike humans. It is something that doesn’t feel hunger, doesn’t get angry, doesn’t get tired and is not prone to making mistakes.

Science-fiction writers, unshackled by constraints of reality and not bound by the rules of contemporary science, have thought of thinking machines for a very long time. These thinking machines were characterized in various ways, but one of them was to call them intelligent. Not in a human sense, but in an artificial manner. These machines would then be called “artificially intelligent” or considered to have “artificial intelligence”.

The idea behind AI is different from that of a mere computational device. A computer, as a computation machine, divides each and every task into more manageable, smaller tasks by representing it in the form of “0s” and “1s”, which effectively means that anything which can be expressed in “black and white” can be handled by a computer. The idea behind AI is that of a machine which is more human-like, a machine which can handle ambiguous situations as well as humans can, and/or which can also look at situations and possible outcomes in a probabilistic manner to handle the “grey” areas.

Any entity, whether it is a machine, an algorithm or a robot which can think and take decisions independently on objective and subjective matters, without the guidance of humans, can be said to have AI. Anything that doesn’t reside in the human body or has not developed as a result of a natural evolutionary biological process over millions of years, and was developed in laboratories by scientists and engineers, would qualify as AI. Under this definition, AI can function like humans in tasks where the directives are not always straightforward and mathematically definable. Furthermore, the development of cognitive artificial systems which are able to perform jobs typically executed by humans, such as two-way conversation or instant decision-making, are all part of the development of AI.

Many of the developments which brought us to the current state of affairs in AI are relatively recent. It would be pertinent to highlight the role of World War II (WWII) in this. The all-around scientific development during WWII also included a lot of interest in computation and statistical thinking from the political and military leaders. For example, the interception of secret communication code and code-breaking was a big war effort on both sides, and there is sufficient evidence available of instances where the Allied Powers used it very effectively to gain an upper hand over Axis Powers such as Germany.¹⁰

Where all of this began less than 100 years ago: Rossum’s robots

The word “robot” can be traced back to a 1920 play, R.U.R. or Rossum’s Universal Robots,¹¹ by the Czech writer Karel Capek. In Czech, robota means serfdom. Rossum’s robots were synthetically produced, servile, flesh and blood creatures. The robots willingly agree to direction from humans and gradually replace humans in various activities. Robots become cheap, are readily available everywhere, and become indispensable as a source of cheap labor. As the play progresses, however, Rossum’s robots become discontent, stage a revolution, and kill all humans. The play ends with some hope though as a few robots start displaying human-like emotions, and there is some hope that they might evolve into a human-like form. The play itself would have been forgotten, but has become immortal because of the word robot it contributed.

Isaac Asimov coined the term robotics in the 1940s while writing his hugely popular robot stories.¹² Asimov’s Three Laws of Robotics, serving as the basis of robot-ethics, became well known among science-fiction readers. As introduced in Asimov’s 1942 short story, “Runaround”, the three laws read: (1) A robot may not injure a human being or, through inaction, allow a human being to come to harm; (2) A robot must obey the orders given to it by human beings except where such orders would conflict with the First Law; (3) A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws. These laws provided a framework for the action of robots in Asimov’s stories, and are so natural that they were referenced by other authors and even in other genres.

Some of these fictional ideas assumed a semblance of reality as the British mathematician and the famous WWII code breaker, Alan Turing,¹³ wrote a paper in 1950 in the journal Mind which introduced the Imitation Game¹³. In his paper, Turing developed a method to find out if a machine can exhibit intelligent behavior indistinguishable from that of a human. Turing really wanted to investigate if a machine could, in principle, think. The paper starts with the words, “I propose to consider the question, ‘Can machines think?’”.

However, steering clear of the ambiguity that the concepts of thought and machine could lead to, Turing instead devised a more practical test—the Imitation Game—through which an interrogator could determine whether the responder was a human or a machine. The machine’s job, in this game, is to try to fool the interrogator into thinking that it is a human. Turing suggested that the machine could convince a human about 30 percent of the time in a five-minute conversation. Turing’s paper did not have much of a role in the practical development of AI. However, the philosophical impact of the idea was significant; one could not dismiss the idea of thinking machines so easily.

Turing did not think of his paper in a vacuum. He was a member of the Ratio Club,¹⁴ a group of young upcoming but really smart mathematicians, engineers, psychologists and physiologists, who used to meet regularly to discuss issues in cybernetics. Favoring the young and the daring, professors were banned from the group. Any member attaining professorship in his professional academic life was expected to immediately resign from the group. The group discussions tended to be extremely thought provoking. Many of the members went on to contribute significantly to their fields.

Among the members was Horace Barlow, Charles Darwin’s great-grandson, who made significant contributions to the neuroscience of vision. Albert Uttley, another member of the group, researched radar during WWII. Thomas Gold, an astrophysicist who helped solve the mystery of pulsar, a type of rotating neutron star, was also a member of the club. Machine intelligence was a frequent topic of discussion at the group meetings, and it is very possible that Turing was motivated to write his paper following discussions with the Ratio Club members.

The Dartmouth workshop: Assembly of the finest minds

A number of developments after WWII led to an interest in automated and intelligent systems. Perhaps...