Artificial Intelligence: Do we really need to fear of it?
In the early spring of 2017, I read some books on AI, including the book “Homo Deus: A Brief History of Tomorrow” by Yuval Noah Harari. Here, I’d like to write down my impression of this book, as well as the film “Humans” and the TV play “Westworld” I watched before.
For thousands of years, no matter the Chinese people in the 20th century, or the Indians in the Middle Ages, or the Egyptians in ancient times, all humans faced the same three problems-- famine, pestilence and war, which always were our biggest concerns. But at the beginning of the third millennium, humans became aware of one surprising thing. Though most people seldom consider it, in the past several decades, we successfully controlled famine, pestilence and war. Though such problems are not totally solved, they have turned from “incomprehensible uncontrollable natural forces” in the past into “challenges that we can cope with”.
Next, humans will face more challenging new problems: eternal youth, lasting pleasure and AI. Yuval Noah Harari, author of the book “Homo Deus: A Brief History of Tomorrow” and an Israeli historian, thinks the future human society will be highly polarized. Most people will no longer have military and economic value. To most modern work, 99% human characters and abilities will be needless. As algorithms push humans out of the job market, wealth and power will likely be concentrated in the hands of several elites. The social system still will regard some unique individuals as valuable, but those people will be a group of superhuman elites, not common people. They will form a privileged class made up of very few people. Those superhuman will be able to make the most important decisions in the world.
In the film “Humans”, one of the products of AI—robots in workplaces and homes are set as the context. In the film, robots can learn by themselves, have feelings and emotions caused by their awareness and experience, have new reactions and behaviors not set in the instruction book or program, which make humans astonishing and alarming.
Similarly, the recently popular “Westworld” describes a game field designed by humans in the wild valley of the western United States. In the end, robots win and humans lose, as Shakespeare wrote in “Romeo and Juliet”: “these violent delights have violent ends.”
Though most content of “Homo Deus: A Brief History of Tomorrow”, “Humans” and “Westworld” is fictional, and many people are not aware of it or haven’t perceived it, our world is changing quietly and subtly. Facebook’s founder Mark Zuckerberg and his Chinese American wife Priscilla Chan have promised to donate USD 3 billion in the next decade for disease research, which means in the future it will not be impossible for humans to live 100 or even 150 years; longer lives will give humans more time to finish more missions or research. At the beginning of 2017, the Master of DeepMind won all human chess players by 60:0 and finished the one-week open test online. On the evening of Jan. 4, 2017, before playing the 60th game with Gu Li, the Master revealed it was the latest version of the AlphaGo that defeated Lee Se-dol in the man-machine competition last year. The Master’s deputy was the AlphaGo Team’s Dr. Huang Shijie. At the Rivers Casino in Pittsburgh, USA, an AI called Libratus defeated four top-class human players in 120,000-hand Texas Hold‘em Poker games and won a prize of USD 200,000. In China, there are battles between Baidu Brain and super human brains. Google, Baidu and Uber are trying and practicing in the unmanned driving field. Some of these are not completed or put into use yet. But from the birth of the world’s first computer (Atanasoff–Berry Computer in 1937) to today’s mobile internet and machine learning, only 80 years have passed. Today, the speed of computer development is far faster than the speed in the past decades. Therefore, it’s foreseeable in the near future we will have more and more products relying on machine learning; our ideas and demands for new technologies will have great changes.
According to the statistics published at Guangdong Province’s Economy and Information Technology Conference in 2017, Foxconn has applied 40,000 robots in its factories around China. In addition, data shows in 2015, the sales of industrial robots in China were 68,459 robots; In Guangdong in 2016, the output of industrial robots grew 45.2%; 22,000 robots were newly applied; the total number of robots in use was 60,000. The biggest challenge to humans brought by AI and robots is mass unemployment. Replacing human workers with robots in simple repetitive work is the trend. Maybe in the future, robots will become the main labor force. However, even if AI will appear more and more in our daily life and workplaces, the idea that AI will conquer or defeat humans is still too fictional. Though AI is more powerful than humans in computing and memory, it cannot surpass humans in terms of emotion and creativity, etc.
