The power of the Machine
© Colin Anderson/Getty
December 2016

The power of the Machine

Winning games, reading files, diagnosing diseases: will robots soon be smarter than people? Will machines be pushing a button someday to turn on humankind? Even experts like Elon Musk and Ray Kurzweil warn of the risks posed by forms of “superintelligence”. And, if so, what will become of us?

The human looks at the robot and says: “Human beings have dreams. Even dogs have dreams, but not you, you are just a machine. An imitation of life. Can a robot write a symphony? Can a robot turn a canvas into a beautiful masterpiece? ”The robot looks at the human and asks: “Can you?”

Of course he can’t. The human’s name is Del Spooner – played by Will Smith – the movie is “I, Robot.” It’s not a very recent release, but its theme is all the more appropriate for our times – the fear that machines intend to surpass us with their artificial intelligence and extinguish us because we’re nothing but pests that destroy the ­planet. Interestingly enough, the three robot laws that prohibit machines in the film to harm humans were written by science fiction author Isaac Asimov in 1942, so the fear of computers is old as computers themselves.

Facts catch up with forecasts

The prediction that computers will be smarter than we has been made by Ray Kurzweil. The American is an author, futurist and Director of Engineering at Google – and his priority at the moment is looking at “singularity,” which he defines as a fusion of humans and machines that will allow us to live forever. Now this could be dismissed as a ridiculous idea if the man hadn’t previously predicted the existence of cell phones, self-driving cars and intelligent weapons systems. Kurzweil himself no longer considers his prophecies to be as radical as in the past. This may be due to the fact that the development of artificial intelligence is taking place at such a rapid pace that it’s impossible to keep track of his predictions.

  • 100 B

    nerve cells are contained in the human brain. It weighs about 1,400 grams – and incurs pretty low maintenance costs. About 3,000 euros have to be spent on food while “air and love” are free. By contrast, Sunway TaihuLight, the world’s fastest computer, incurs annual electricity costs of eleven million euros. In return, it performs trillions of mathematical operations per second, compared with the “mere” ten billion by our brain. Another difference: we crash clearly less frequently – and when we do it’s typically by intent.
  • $100.000

    is the prize money awarded for passing the Turing Test incepted by U.S. sociologist Hugh G. Loebner in 1991. The challenge is to create a machine that convinces a tester that it’s a thinking human being. Whether the chatbot “Eugene Goostman” in 2014 was the first subject to pass the test continues to be controversial. The chatbot passes itself off as a 13-year-old Ukrainian who, among other things, owns a guinea pig and likes Eminem.
Artificial intelligence is encroaching on the domains of human smartness

Machines by now are not only better at playing chess but also at Go, which is an even far more complex game. In March, Google’s computer program AlphaGo beat Lee Sedol, the board game’s champion. “We’re in a complete state of shock,” the South Korean commented following his defeat. The machine’s victory was enabled by Deep Learning, artificial neuronal networks. Thanks to Deep Learning, artificial intelligence is increasingly surpassing us in domains that, up to now, have been deemed to be realms of human smartness.

Algorithms recognize traffic signs and disease patterns, and distinguish cancerous from healthy cells, which means that not only taxi drivers or excavator operators will be replaced by self-driving vehicles – at some point in time, we’re no longer going to need doctors or journalists either. Even today, financial news are written by programs and some U.S. law firms use AI systems that plow through files with greater accuracy than any legal assistant. At the same time, it’s appropriate to say that industrialization and mechanization have always devoured jobs – and produced new ones.

Robotic referees heading for the Olympics

Even the job of the referee is threatened with extinction. The Fujitsu Corporation has jointly developed an automated referee with Japan’s Gymnastics Association that judges rhythmic sports gymnasts. The system is expected to be used in the 2020 Olympics in Tokyo. Why not? An electronic brain can’t be bribed and doesn’t tire.

It’s just that we have reservations about machines because they pass inhuman judgments and have no concept of values. In addition, we fear a loss of power – the discussions about automated driving being a case in point. Plus, we consider ourselves to be morally superior, albeit, of all reasons, because we feel that our personal judgments are the results of a cognitive process, except that artificial intelligence fares better exactly in this discipline.

The power of the Machine© Colin Anderson/Getty
Humanized machines

But at least we have feelings, feel love and cultivate relationships! The only question is what we mean by that. Even years ago, the U.S. company True Companion presented a sex robot called Roxxxy that listens, talks, senses touch and whose personalities can be selected according to the owner’s whims – Roxxxy may be shy on some occasions and bold on others. Roxxxy was never particularly eloquent, but she arguably didn’t have to be. For Ray Kurzweil, there’s no question that we will be mating with machines someday, albeit primarily in virtual forms and with nanobots in our bodies sending sensory signals. ”Sexual pleasure,” Kurzweil once wrote “is an invention produced by the brain itself – just like humor or rage.”

Those familiar with the HBO series “Westworld” have to rethink the meaning of the term empathy with respect to an individual and a machine anyhow. Based on the novel by Michael Chrichton and the same-named movie, “Westworld” tells the story of a western pleasure park populated by avatars in which humans treat machines like animals. As a viewer you feel bad – and empathize with the machines.

  • 85 %

    is the probability of a computer recognizing that a human being truly feels pain. In a study conducted by the magazine “Current Biology,” a computer program checked videos of faces distorted by pain. Amazingly, only 50 % of the judgments made by humans were correct. By the way, Marian Barlett, who was in charge of the study, sells an app that by means of a smartphone camera decodes facial expressions to enable companies to better understand customer responses.
  • 24

    hours were spent on the internet by the self-learning chat program Tay before Microsoft had to remove it. Tay was supposed to learn how 18- to 24-year-olds communicate and join the conversation. However, within a short period of time, Tay drew Hitler comparisons, saw Donald Trump as the last resort and denied the Holocaust even before Microsoft was able to pull the plug. The problem was that, without major reflection, Tay inferred its responses from what other Twitter users had written.

Who decides that robots have no dignity? In the debate about the distinction between humanoids and androids, the argument is essentially put forth that a computer has no body and hence feels no fear, no disgust and no happiness. Furthermore, machines, unlike humans, could not create but only replicate. No matter which way you look at it, the five-minute science fiction film “Sunspring” is the first work to feature a screenplay that has been entirely written by an algorithm.

Admittedly, the film primarily stands out due to its bizarre dialogs – but viewers of man-made television are certainly prone to suffer feelings of confusion as well. Plus, the number of original motion pictures isn’t very long either. Instead, we watch prequels, sequels, spin-offs or remakes. So much for human ingenuity.

Prominent warners

Yes, perhaps we have reason to be afraid. Astute people like the Swedish philosopher Nick Bostrom, Stephen Hawking, Bill Gates or Elon Musk indeed are and issue public warnings of an intelligence that is uncontrollably taking on a life of its own. However, the question is whether it will even be interested in us, which is hard to imagine. And what will it find? Humans staring at smartphones. In this light, we’ve already succumbed to the digital.

Wiebke Brauer
Author Wiebke Brauer
Although Hamburg journalist Wiebke Brauer is afraid of artificial intelligence and its consequences, she has no problem shifting parts of her brain to her iPhone. As a result, she remembers only two phone numbers – as opposed to countless ones remembered by the device.