Hello, My Name Is Human
Progress throws new challenges at us and humankind has demonstrated that it’s unprepared for the future. Our civilization made stunning breakthroughs in the 20th century but, with each one, acquired new fears and self-doubts. Social injustice, ecology issues, psychological concerns, and ethical questions — all caused by the industrial revolution and technical progress. The most disturbing topic, arguably, is whether machines will take our places at work, home or in society. The process has already begun but we can only imagine how it will end.
People first started losing their jobs to technical progress at the end of the 19th century, when manufacturers broadly implemented production lines. Since then, we have considered machines a threat to underqualified workers, at most. However, that threat has grown throughout the century. Before we know it… Boom! Machines have replaced factory assemblers. Bang! AI is capable of developing software for itself.
Machines are already better than humans in many fields. Luckily for us, expansion doesn’t happen that fast. Some jobs are still inaccessible to machines and the situation is unlikely to change soon. In what ways are machines already superior to people and what can they learn from us?
Let’s begin with some examples of machines’ supremacy.
Work in Deadly Environments
Humans are weak. We can’t survive in many places on Earth. Our kind isn’t adapted for the rough conditions of the planet we live on, so to explore those areas, people have created machines. Robots, drones, rovers, self-driving spacecraft, pre-programmed machinery — they all examine the planet’s oceans, mountains, atmosphere and near space. We went a step further and launched machines to observe our solar system and its planets for us.
In 2017, the European Space Agency (ESA) sent a research spacecraft to the Churyumov-Gerasimenko comet. The probe, Rosetta, landed on the comet’s surface and took pictures of it. In October 2018, in cooperation with its Japanese colleagues, ESA launched two satellites to explore the planet closest to the Sun, Mercury. It’s hard to imagine deadlier places for humans, therefore those are machines-only missions. Probably, one day, we’ll go to Mars or down to the bottom of the Mariana Trench but for today, robots do what we simply can’t.
Calculate and Process Data
Naturally, machines can perform the rather physical conditions of work better than humans, but machines are also better in traditionally human tasks. We’ve gotten used to thinking that humans are superior to anything else in this world when it comes to computing and processing data. Machines, though, are able to collect and handle a massive amount of information, then self-learn to manage it more efficiently. Computers process billions of operations per second and we will never surpass that.
In 2017, the startup Recursion Pharmaceuticals began employing AI that understands the anatomy of cells to discover new uses for existing drugs. Other researchers are sure that AI can wrestle with the amazing complexity of quantum systems containing billions of particles. Finally, what terrifies us most, Researchers at UCL, the University of Sheffield and the University of Pennsylvania performed an experiment in which artificial intelligence predicted the judicial decisions of the European Court of Human Rights with 79% accuracy. How far are we from AI-ruled courts with the authority to decide a man’s fate?
We’ve created artificial intelligence to organize astonishing quantities numbers and data, but computers aren’t perfect. Developers enhance them every year. However, it’s hard not to notice the evolution of the algorithms, which are getting smarter with frightening speed.
Who could’ve imagined that one of the best ways to test machines would be through games? Not football or baseball but intellectual games, such as chess or Go. Computer-vs-human chess competitions began in the ’50s and ended in 1997 when the IBM chess computer Deep Blue beat the best chess player in the world, Gary Kasparov. Since the beginning of the 21st century, the majority opinion has been that humans will never again beat machines in chess and nowadays, you can freely download chess programs that will easily crush the best of the best.
The game Go is much harder for computers than chess. After the first two moves of a chess game, there are 400 possible next moves, whereas Go has about 130,000. AI developers mentioned that the number of all possible combinations in Go is greater than the number of atoms in the universe. Developing an AI capable of defeating man took a long time, but it all ended with a three-match series where the machine prevailed against the world’s best Go player. That game was our last hope! Many thought that it was too hard for machines. Has the last stronghold fallen?
However, some things are still incomprehensible to machines and nobody can predict when that fact will change.
Cope with Human Emotions
What machines are incapable of is human nature. The algorithms may imitate empathy but it remains artificial. True empathy assumes a significant overlap in experience between the subject of the empathy and the recipient. Although some companies are trying to develop machines with the ability to offer human emotion, they haven’t succeeded yet.
Where computers have advanced is in passing the Turing test. However, will robots ever be able to convince us of their authenticity? Who knows. The uncanny valley effect still exists.
Work with Language
In everyday life, you can book a hotel room using your smartphone assistant. Google has presented the amazing Duplex feature, a voice-assistant technology, which impresses and terrifies at the same time. How does Google Duplex convince you of its humanity? It speaks to you with very natural and fluent language. English, of course.
There are two factors for machines when processing language: synthesis and recognition. Text-to-speech technology exists and is developing quite quickly. If machine voices in ’80s movies sounded very mechanical, then modern robots speak in relatively natural voices. The problem is, before it can speak, the machine should recognize exactly what it needs to vocalize. Language recognition is an especially complicated process. Yes, Google Translator will understand you and help you communicate with a foreign friend, but it can hardly replace professional translators. Aspects like style, semantics, grammar (which is complex in many languages), culture, and context adaptation are beyond any of today’s machines.
Thinking Critically and Creatively
“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?” These are the questions of Will Smith’s character in the 2004 movie I, Robot.
Machines imitate existing painting templates, write poems, even music and movie scripts, but this isn’t creativity. All these algorithms were operating in conditions under human control. They were pre-programmed for those tasks. We can’t regard the products of their creativity in the same way as art by Vincent van Gogh, Tom Yorke, or Quentin Tarantino. They are different. Moreover, only people can consider anything a masterpiece. Art is still a purely human privilege.
We are creating and developing machines that, probably, will rise against us one day. Although may people lose their jobs and face new issues, they gain much more in return. We have nothing to fear as long as we set the rules, and until machines learn to dream and reproduce themselves, everything will remain under control. Cogito, ergo sum is still our strength.