The Global Intelligent Machine

Teun Koetsier

The Machine Stops is a story from 1928 by Edward Morgan Forster. It is about a machine that fully controls life on Earth. Mankind lives underground because the surface of the Earth has become uninhabitable. On the surface, where the air is hard to breathe, only a few Homeless, as the machine calls them, succeed in surviving. The protagonist of the story is a young man called Kuno. He feels that there must be more to life than full obedience to the rules of the machine, and so he rises up against it. The resulting major and irreparable malfunctions bring the machine to a halt, and cause many deaths. Kuno and a few survivors are left on the surface of the Earth, where, together with the Homeless, Kuno will try to rebuild a civilization in harmony with nature.

Forster’s tale is one of several similar stories that appeared in the first half of the 19th century expressing the feeling that technology separates man from nature, that it degrades man. Aldous Huxley’s Brave New World, written in 1932, is another famous example. The stories can be seen as elegant literary criticism of things that were actually happening, but they were obviously fiction. For good reasons, Huxley set his story in the year 2540, hundreds of years in the future.

Things have changed. Right now, at the beginning of the 21st century, the technology needed to build one huge machine that completely controls a society or even the world comes into view. In this development, the technologies of connectivity and artificial intelligence play a central role. Even in 1970, when Alvin Toffler wrote Future Shock, the fantasies of Forster, Huxley, and others still seemed pure science fiction. Yet Toffler realized that the industrial societies as we knew them were rapidly changing. Ten years later, in The Third Wave, Toffler discussed the “intelligent environment” and predicted the rise of intelligent networks.

Information Machines

A machine used to be a device that enhances our possibilities to intervene in nature. I call such artifacts—like windmills, steam engines, cars, and airplanes—production machines. They dominated technological development in and after the Industrial Revolution. The result was the large-scale mechanization of man’s physical interaction with nature and a complete transformation of society. Agricultural societies became industrial societies.

Nowadays computers dominate our life. Computers are not production machines; they are information machines. Instead of intervening physically in reality, they process information—and they do so quite efficiently. Production and information machines can be combined in hybrid machines. Robots are hybrid machines: sensors collect information, on the basis of which a computer calculates the moves that are executed by production machinery. We have clearly entered the age of robots. The number of industrial robots worldwide is around 2.5 million at the moment, and it is growing. Military robots are now a standard part of the defensive arsenal. Service robots have moved into areas like removal of land mines, volcano exploration, deep-sea exploration, extraterrestrial exploration, construction works, agriculture, mining, cargo handling, cleaning, and helping the elderly. Completely automated vehicles have appeared on the roads, and while the public might not yet be ready for it, pilotless flying is a possibility.

Indeed, the development of computer technology has completely transformed society. Industrial societies have become information societies in which—quite in accordance with Toffler’s views—all our institutions are shaking under the hurricane impact of the accelerative advance of technology.

The Global Intelligent Machine

While in the past machines were always local, we are now living in an age of connectivity, and this has profound implications. Automated connectivity has led to machines that extend over large areas—a factory, a city, a country, even the globe. The internet is very different from, say, a steam locomotive, but it is a complex coherent device operating on the basis of the laws of physics, designed to perform a task. That means that it is a machine—not a production machine, but a huge global information machine comprising a network of computers.

“

While in the past machines were always local, we are now living in an age of connectivity, and this has profound implications.

”

Originally, the internet did not collect the information it processed. It merely processed the information that we human beings put into it. That has changed. The internet has become the Internet of Things, which is the internet combined with billions of sensors that automatically gather information that is subsequently processed and stored. The Internet of Things is a huge global information machine and it, too, is growing. As more sensors are added, we are again witnessing a new development. Artificial intelligence software that tries to understand the data is being incorporated more and more into systems. The Internet of Things is getting smarter. Because production machinery is also increasingly being hooked up, the Internet of Things is no longer merely an information machine, but is becoming a gigantic hybrid machine. For this global machine I coined the name Global Intelligent Machine (GIM).¹ GIM has sensors to register the world, brains to process the information, and production machinery to intervene in the world.

Industrial robots, service robots, self-driving vehicles, and in general all smart machines that will be built in the future, will be connected to GIM. Because there is not one central control and there are many users, GIM is a robotic machine, but not a robot, although in principle it could become one. Robots extending over considerable areas like automated ports, factories, and farms already exist.² Singapore’s prison system, for example, is moving toward automating work processes in prison. It calls the concept “prison without guards.”³ Jörg Goldhan, of ETH Zürich, argues that robots will make doctors obsolete in the foreseeable future, allowing hospitals to significantly save on costs.⁴ And indeed in 2017 a robot passed China’s national medical exam, scoring 96 points higher than the required 360.⁵

“

The idea that maybe even before the end of this century machines will be able to imitate human beings in such a way that we will be inclined to accept them as our equals may sound bold, but it is no longer incredible.

”

Are Chess-Playing Robots Really Playing Chess?

Imagine an unmanned spaceship floating somewhere in space many light-years away. On-board robots are playing chess. Incessantly the robots calculate their next moves, and when the game is over, they start again. All contact with Earth was lost long ago, and nobody is watching the game. Are these robots really playing chess? Of course they are, you might say, if you adhere to realism in philosophy. Trees in the forest don’t stop existing simply because nobody sees them.

However, imagine that a wave on the beach casts two shells at my feet. The next wave adds three shells. Is the sea calculating 2+3=5 here? This sounds strange, but don’t worry. The events on the beach are accidental and essentially need an observer to be interpreted as 2+3=5, while the robots were designed to play chess. The sea does not calculate, but the robots do what they are designed to do, irrespective of any observers. Somebody imagined them playing chess, designed and built them, and that is, from a realist point of view, sufficient to say that they play chess.

I am a realist. Yet what is taking place on the spaceship seems quite pointless and not at all a game. The robots go through the moves and, strictly speaking, perhaps they are playing chess, but I would argue that they are not really playing chess unless they also possess qualities like being able to enjoy the game and, more generally, by being aware of themselves and of what they are doing.

At present robots do not have these qualities, but this may change. In 2005 Ray Kurzweil made the daring prediction that within decades machines would become superior to human beings in every respect, not only in problem-solving skills, but also in emotional and moral intelligence.⁶ This is the event he called the “Singularity.” The idea that maybe even before the end of this century machines will be able to imitate human beings in such a way that we will be inclined to accept them as our equals may sound bold, but it is no longer incredible. The technological development is unbelievably fast and there are no signs that it will slow down. The question is when the imitation, the replica, will be accepted as real, as equivalent to us.

“

Those who assume that in every wave of automation, new jobs are created elsewhere in the economy, forget that in the past there were always lots of things machines could not do. We are now talking about robots that can do everything a human being can do.

”

It may happen without us even noticing it. Already, important decisions that in the past were made by humans are left to computer programs. In the US, parole boards sometimes rely on software that calculates whether an inmate should get parole. There also exist programs that analyze traces of DNA from crime scenes. The way this software works is a trade secret.⁷ The possibility that everywhere in society superior software will gradually replace human beings is real.

In the history of technology mankind has again and again entered uncharted territory. Now it may be more than business as usual. Basically, two things define a human being. We process information, and we interact physically with nature. After the Industrial Revolution, most if not all aspects of our physical interaction with nature were mechanized. Moreover, computer technology has made it possible to mechanize many aspects of our information processing. It seems we are on our way to mechanizing ALL aspects of our human existence. Such a thing has never happened before. We seem to be nearing a crucial limit of some sort.

The Impact

Currently, in spite of its complexity, your computer is an object and you can do with it whatever you want. You can sell it to someone else, and even if you demolish it in a fit of temper it is unlikely that you will be accused of serious wrongdoing. It will become very difficult to treat robots or robotic machines like GIM merely as objects once they possess emotions and are conscious of themselves. It seems inevitable that they will claim rights and subsequently force the world to respect those rights. A pessimist might say that once the superior intelligence of GIM also includes emotions, moral insight, and self-awareness, we will have created Forster’s machine. This may very well happen before the end of this century.

The optimist must assume that we will succeed in controlling the development in such a way that GIM turns out to be benevolent. Already we are struggling to adapt to the relentless pace of technological development. The undesirable effects of the ease by which messages can be spread over the globe are obvious. The internet does not distinguish between entertainment and information, between truth and falsehood—even while fake news attracts more attention. The traditional sources of reliable information—science, critical journalism, and the state—are competing with stories produced by idiots, lunatics, and adversaries of Western civilization. Cybercrime is already a major problem and will remain so. The rise of GIM, the Global Intelligent Machine, makes the risks much bigger. Without proper security measures, the production machinery that is part of GIM can, in principle, be manipulated from a great distance by unauthorized entities. For a nice survey of all the risks involved in the rise of GIM, I refer to security expert Bruce Schneier’s aptly titled book Click Here to Kill Everybody.⁸ Those who assume that in every wave of automation, new jobs are created elsewhere in the economy, forget that in the past there were always lots of things machines could not do. We are now talking about robots that can do everything a human being can do. The effect on our economies cannot be less than substantial. GIM is an incredibly impressive machine, but it confronts us also with major challenges.

Teun Koetsier is a historian and philosopher of mathematics at VU University Amsterdam, the Netherlands. His research covers a wide range of subjects. While working on a “Big History” of production and information machines, he realized that what started with simple production and information tools like hand axes and tally sticks seems to lead inevitably to a situation in which all existing modern machinery becomes part of one big global hybrid machine, GIM, the Global Intelligent Machine. See https://www.springer.com/gp/book/9783319965468.

1. Teun Koetsier, The Ascent of GIM, the Global Intelligent Machine, A history of production and information machines, Springer Science Publishers, 2019.

2. The phenomenon of connectivity is related to a third aspect of our species, next to the fact that we intervene physically in nature and process information. From the very beginning man was also a creator of networks, a homo reticulorum. For a sketch of the history of networks from the Stone Age until the present see: Teun Koetsier, From Early Trade and Communication Networks to the Internet, the Internet of Things and the Global Intelligent Machine (GIM). Advances in Historical Studies, 8, 2019, pp. 1–23. doi: 10.4236/ahs.2019.81001

3. https://www.channelnewsasia.com/news/singapore/changi-prison-raises-tech-bar-with-automated-checks-surveillance-10455870

4. Jörg Goldhahn, Vanessa Rampton and Giatgen A. Spinas. Could artificial intelligence make doctors obsolete?, BMJ, vol. 363, pp. k4563, London: British Medical Journal Publishing Group (BMJ), 2018. DOI: 10.1136/bmj.k4563

5. http://www.chinadaily.com.cn/bizchina/tech/2017-11/10/content_34362656.htm

6. Ray Kurzweil, The Singularity is Near, When Humans Transcend Biology, Penguin Books, 2005

7. New York Times, June 13, 2017. See https://www.nytimes.com/2017/06/13/opinion/how-computers-are-harming-criminal-justice.html

8. Bruce Schneier, Click Here to Kill Everybody, Security and Survival in a Hyper-connected World, W. W. Norton and Company, 2018.