Artificial Intelligence: How It Changes the World

By Tatev Ghazaryan. Originally published July 29, 2019 on Enlight

 Is Artificial Intelligence (from now on AI) a gift to humanity or a curse? This debate has long been discussed and continues to remain relevant. The “invasion” of the AI to the human ecosystem is no longer an undeniable fact. Its application areas are many and varied: including healthcare, management, military, financial and economic applications, advertising, the arts, and so on. At the same time, the AI, penetrating the economic, political, and military spheres, is becoming one of the decisive factors of the state’s power, affecting geopolitical processes as well. The pursuit of leadership in the AI field forces countries to invest more and more in the development of the sector, which in turn stimulates the AI race.

What Is Artificial Intelligence?

The AI doesn’t have a universal definition. Although it is due to the emergence of the electronic computing system and artificial neural networks in the 1940s, nowadays it concerns a wider circle of subjects, technologies, and methods. Contrary to the widespread perception of AI, it is not a new science branch. Many of its foundational concepts are based on over 2000 years of accumulated experience in philosophy, logic, mathematics, cognitive psychology, linguistics, and other sciences. However, the AI became a research field after World War II. In 1950, Alan Turing published his article “Computing Machines and Intelligence”, where he proposed the idea of an “Imitation Game”, which was built around the question “Can Machines Think?” Later this idea was qualified as a Turing test, which measured the machine’s (artificial) intelligence. Turing’s experiment became an important component of the AI’s philosophy. The term AI was first proposed and circulated by the American scientist John McCarthy in 1956 as part of a Dartmouth seminar organized by him.

In order to better understand what the AI is, let’s try to clarify its definition. So, the co-author of the textbook, “Artificial intelligence: Modern Approaches”, Professor Stuart Russell of the University of California, Berkeley, defines AI as “a study of methods that allow computers to operate rationally”[1]. In its broadest sense, AI is defined as “the study of calculations that allow us to perceive, judge, and act” or “the automation of intelligent behaviour”. In general, the AI can be defined as: non-human intelligence, which is measured by its ability to reproduce a human’s mental abilities, such as image recognition, object recognition, the semantic understanding of language, experience-based adaptive training, logic, strategy development, the ability to make judgments about phenomena, decision making, and so on.

Referring to the AI classification, it should be noted that currently there are two levels of the separation of the AI: weak (narrow) and strong (general) related to the degree of human capacity reproduction by the AI, however, there is often a third level in literature – the super-intelligence. These levels can also be viewed as a succession of the development of the AI’s generations.

  • Weak AI, which is often called narrow artificial intelligence, is an AI system designed and developed for a particular problem. It can be equivalent as well as exceed human capacity for certain tasks. Examples of such systems are IBM’s Deep Blue chess program, Watson (‘Jeopardy!’), Google’s AlphaGo (go), virtual personal assistants such as Apple’s Siri, or any other specialized automated systems that work within human capacity. All currently existing AI systems are classified as either narrow or weak.
  • Strong AI, which is also known as general artificial intelligence, is an AI system with generalized human cognitive abilities. When a strong AI is presented with an unknown problem, it can solve it without human intervention. It corresponds with the full range of human abilities and can successfully accomplish any mental task that is accessible to man.
  • Artificial super-intelligence or hyper-intelligence is a machine intelligence that surpasses human intelligence in any matter. Often the main fears in predicting the future are connected to the artificial super-intelligence.

Pros and Cons of Artificial Intelligence

The opinions circulating about the potential impacts of the AI are diverse and stand out with both optimistic and pessimistic tendencies. The fact that it helps to reduce the probability of errors and increase accuracy can be secluded from the advantages of the AI. It can be used to perform various dangerous missions, such as for research and study purposes of the universe, the ocean floor, mines and other places inaccessible or restricted to humans. The self-driving vehicles operated by the AI will help to reduce the number of car accidents. In addition, AI has widespread use on a daily basis. The smartphone is the most common example of how we use artificial intelligence. When we take photos, the AI algorithm identifies and recognizes a person’s face and mentions people when posting pictures on social media. AI is widely used in data management in financial and banking systems. Nowadays it can detect frauds in card systems. Developed organizations use «avatars», which are duplicates or digital assistants, and actually, interact with users. The lack of emotions allows robots to judge logically and make rational decisions. The machines, unlike humans, can work continuously, thereby increasing the efficiency of work. AI can replace a human in repetitive, boring, and monotonous tasks by acting faster and performing multiple functions simultaneously. The widespread use of AI can also be seen in medicine. The doctors evaluate the patient or his health risks with the help of the AI. It also provides information on the side effects of various drugs. One of the reasons for the growing role of AI is huge opportunities for economic development. According to project estimates done by PriceWaterhouseCoopers in 2017, AB Technologies can increase world GDP by $ 15.7 trillion (14%) by 2030.

Along with all these positive aspects, the AI also suggests problems. First of all, AI implies big expenses, resources, and time to build, rebuild, and reconstruct it. In addition, many ethical and moral issues related to AI are being discussed today. The AI is devoid of creativity and operates only within the boundaries that a human draws for it. Another important concern is that although the AI reduces the likelihood of human error, the problem may be in the AI code itself, which is ultimately written by a human. The other major issue that the widespread use of AI can cause is unemployment. Recent research shows that by 2025, about 50% of current jobs will be eliminated, as will many companies and industrial brunches. Finally, people may be overly dependent on the AI, which can lead to a decline in an individual’s mental capacity. The appearance of the machines in the wrong hands is also a cause for concern for mankind, and scenarios of “machine rebellion” are also considered in the distant future.

When speaking of a strong AI, Stephen Hawking mentioned: “It would take off on its own, and re-design itself at an ever-increasing rate, and humans, who are limited by slow biological evolution would be superseded.” Meanwhile, futurologist Ray Kurzweil believes that in the future, humans and machines will be able to survive together in a world where machines enhance human capacity while maintaining the norms of coexistence. Henry Kissinger, in his 2018 “How the Enlightenment Ends” article, expresses concern about the rapid growth of the AI, finding that human society is not yet ready for such changes. He also criticizes the AI’s influence on US election campaigns by talking about the ability to target micro-groups, especially through social media. If AI has an opportunity to influence the electoral process in democratic countries, then in authoritarian regimes it can strengthen public control [2].

The Military Adaptability of Artificial Intelligence

In September 2017, in front of a group of Russian students and journalists, Vladimir Putin announced: “Artificial Intelligence is the Future; the one who will become the leader in this field will rule the world.” Three days later the founder of SpaceX and Tesla Elon Musk went farther by tweeting: “Competition for supremacy in the field of artificial intelligence is likely to be the cause of WW3.”

Like many technologies, the AI has hidden military potential. AI applications are already used in the areas of defence and reconnaissance: having their own counterparts in civil life, such as logistics, planning, analysis, transit, data analysis with the support of the AI. These applications are different from those used in a war. They, as a rule, are divided into two groups; which greatly influence the operational level of the war and which also influences the strategic level. At the operational level, the AI can have a major impact on achieving tactical goals. At the strategic level, it can have a significant impact on the scale and chances of war, on escalation and de-escalation, and therefore on political decisions on strategic stability and deterrence.

Nowadays there is a growing interest in self-propelled weapons, and governments are ordering organizations and research centres to develop them. But there is a serious problem connected with deadly autonomous weapon systems or so-called “killer robots”. These weapons can select and attack individual targets without human control. If such systems were to be created and implemented, then such developments would have a major impact on the conduct of wars and after gunpowder and the atomic bomb would be considered the third revolution in the war. The automatic target selection and destruction function can be applied to various platforms, such as battle tanks, ships, planes. Deadly autonomous weapons cause a number of legal, ethical, and security problems. Despite this, the United States, Russia, Israel, and South Korea are among the countries that have opposed UN efforts to ban deadly autonomous weapons.

Artificial Intelligence Brings Forth New Geopolitical Tendencies

The rapid growth and penetration of the AI into economic, political, and military areas makes it an instrument that affects the power of states, thereby also affecting geopolitical processes. 2018 has been a turning point in terms of evaluating the geopolitical impact of the AI. New tendencies of global transformation have emerged that can be conventionally called AI nationalism and AI nationalization.  These tendencies were further emphasized by the fact that in developing their national strategies in the field of the AI many developed countries began to change their attitudes towards liberal principles at the same time. First, instead of global cooperation, global division of labor, the introduction of new platforms, and the exchange of talent, countries (China, Russia, the UK, France, Israel, and South Korea) emphasize the nationalism of the AI, which proclaims their own country’s economic and military interests as the core of its AI strategy. Second, the AI’s nationalization policy assumes the following priorities: engagement of public and private resources, coordination of the pace of implementation of the AI innovations, reorientation of strategic goals: to support the country’s economic growth, geopolitical and military advantages on the international platform.

Some geopolitical forecasts suggest that as a result of the intensification and continuation of these tendencies, the world will undergo significant changes in the near future, and international relations will be determined by the following factors:

  • The AI race between leading countries, which will determine and guide the further development of AI technology. In this respect, the “digital empires” of the US and China stand out.
  • The neo-colonialism of the AI in the relationship between the AI leaders and the countries that are out of the competition [3].

The Role of Artificial Intelligence in Countries’ Strategic Plans

The countries are increasingly aware of the economic, strategic, and military development of the AI. In recent years, France, Canada, China, Denmark, the European Commission, Finland, India, Italy, Japan, Mexico, the Nordic and Baltic countries, Singapore, South Korea, Sweden, Taiwan, the United Arab Emirates, and the United Kingdom have published their strategies for promoting and developing the use of the AI[4]. These strategies relate to education, research, digital infrastructure, the public service sphere, ethics, and other fields. However, not all countries can strive to be leaders in the field of AI. This is more a problem of identifying and building competitive advantages, to meet the specific needs of the nation. Some countries focus on scientific research, others on talent and education, the introduction of the AI in administrative work, and so on.

Among the players seeking global leadership in this area are the USA and China. In July 2017, the State Council of China published the “Plan for the Development of New Generation Artificial Intelligence”, which includes China’s goal: to become the core hub of innovation in the AI sector by 2030, creating a $ 147 million domestic AI industry, and becoming a world leader. 50% of China’s economic activity can be automated, contributing to the rapid growth of the narrow AI. It has also incorporated AI and digital technologies into its geopolitical strategy. Its “Belt and Road” initiative of construction of infrastructures linking Asia, Africa, and Europe since 2016 has launched a digital component within the “Digital Belt and Road” project. The latest achievement of the program in February 2018 was the launch of a new international technology centre in Thailand for the “Digital Silk Road.”

In his turn, in February 2019 the President of the United States Trump signed a decree, which confirms the importance of leadership in the field the AI “For the sake of ensuring the economic and national security of the USA and shaping the global development of the AI in accordance with national values, policies, and priorities.” This decree sets out the American ABI initiative, which is based on five principles:

  • invest technological achievements in the field of the AI in federal management, industry, and academia to promote and ensure scientific discoveries, economic competitiveness, and national security;
  • develop technical standards and reduce the barriers of the safe testing and deployment of the AI technologies;
  • educate the American workers in the development and application of AI technologies;
  • promote public confidence in AI technologies and protect civil liberties, privacy, and American values;
  • contribute to the international environment that supports the American AI researches and innovations, and opens up markets for the American AI industry, at the same time protecting the technological advantage in the field of the AI and protecting the important technologies of the AI so as not to fall into the hands of strategic competitors and rival nations.

It should be noted that, despite the leadership of the USA in the field of AI, today it has no official AI strategy and in this sense is lagging behind the 18 countries that currently have such strategies.

Relatively small players in this area, such as France, the United Kingdom, and Canada, also have ambitious strategies and are making significant investments in the AI field. Other countries specialise in a specific sector of the AI, following a low threshold strategy. India, for example, wants to become an “AB garage” specializing in creating applications specific to developing countries. Poland studies issues related to cyber-security and military use. The United Arab Emirates government launched its AI strategy in October 2017, by creating the world’s first artificial intelligence ministry, whose primary purpose is to use the AI to improve productivity and efficiency [5].

The AI Race

Given the dynamics and potential of the development of the sector, in the not too distant future, it has a tendency to lead to the AI race. To assess what global players can take part in this race, four main factors need to be taken into account.

The first one is data accessibility, which is necessary for machine learning and nourishing AI. In this area, China has a competitive advantage over the United States with more than 740 million Internet and 1.4 billion mobile users, with 290 million and 416 million Internet and mobile users respectively from the US side.

Although China benefits from the competitive advantage of available data and predicts that by 2030 it will hold 30% of the world’s data, due to the stricter rules for their use, it lags behind the USA in terms of attracting talent.

The second criterion is the presence of a favorable ecosystem that allows for the involvement of a larger number of AI companies (such as data usage regulations, privacy laws, etc.). Thus, 40% (1393) of the total number of the AI start-ups in the world (as of 2018 there are 3465) are in the USA and 11% (383) in China, and the third is Israel with 10.5% (362), followed by the UK (7%), Canada (3.8%), Japan (3.1%), France (3.1%) and Germany (3%)[6].

The next important factor, which affects the AI race, is the level of technical skill and competitiveness. Interestingly, the number of graduates in a year in China is twice as high as in the USA. In particular, the number of engineering students has grown incredibly, and in 2015, the top 5 of the highest paid jobs were related to technology, thus attracting more talents. However, despite the growing number of science, technology, engineering, and mathematics graduates, there are 39,200 researchers in the field of AI in China, which makes up about half of the total number of AI researchers in the US (78,700). As for the research work in the field of AI, data analysis has shown that between 2011 and 2015, China has published 41,000 surveys, while the United States – 25,500. However, although China’s publications exceed by volume, it was 34th in industry research citations, which indicates that most of the work did not have the quality that, for example, was in the US research work (4th place). But research done by the Allen Institute shows that China is poised to pass the USA in the number of links to research done in the AI field.

In addition, China, despite considerable efforts over several decades, continues to depend on the USA from the point of view of the production of processors and microcircuits. It annually imports about 200 billion dollars of microcircuits from the US (more than oil). Developing this sector is one of China’s top strategic priorities, for this purpose, China has cut taxes on microcircuit companies and plans to invest about $ 32 billion in this area.

Finally, the availability of sufficient capital is the key to the success of the research in the AI and related fields. In line with this policy, China is making huge investments in strengthening its AI capacity. Thus, the Alibaba organisation invested $ 15 billion in AI research. This figure is comparable to American giants; in 2017, Amazon invested $ 16.1 billion for research purposes in the area. According to a report done by the Wuzhen Institute, during 2012-2016, AI companies of China received $ 2.6 billion in investment financing, which is significantly less than $ 17.9 billion in financing done by the US companies.

Comparing all indicators, we come to the conclusion that although China has competitive advantages in some standards, the United States is still the leader in the AI sector. Nonetheless, it should also be stated that China’s motives are real and clear and should not be underestimated. It has all the opportunities and conditions to be a leader in the field, and at the present, all efforts are directed towards their improvement, so in the near future, there is a great likelihood that the ratio of forces will change in favor of China.

References

[1]http://people.eecs.berkeley.edu/~russell/temp/q-and-a.html

[2]https://www.theatlantic.com/magazine/archive/2018/06/henry-kissinger-ai-could-mean-the-end-of-human-history/559124/

[3]https://geopolitics.co/2018/11/03/artificial-intelligence-as-a-geopolitical-factor/#more-52626

[4]https://medium.com/politics-ai/an-overview-of-national-ai-strategies-2a70ec6edfd.

[5]https://www.cairn-int.info/article-E_PE_183_0105–the-geopolitics-of-artificial.htm#no12

[6]https://asgard.vc/global-ai/

Bibliography

  1. Davis S. Zachary, “Artificial Intelligence on the Battlefield. An Initial Survey of Potential Implications  for Deterrence, Stability, and Strategic Surprise”, 2019
  2. Frank Slijper, Alice Beck and DaanKayser, “State of AI. Artificial Intelligence the military and increasingly autonomous weapons”, PAX, 2019,
  3. Henry Kissinger, “The Enlightenment Ends”, The Atlantic, 2018
  4. Jeffrey Ding, “Deciphering China’s AI Dream”, Future of Humanity Institute, University of Oxford, March 2018
  5. Nicolas Miailhe, “The geopolitics of artificial intelligence: The return of empires?”, Politiqueétrangère, 2018/3 (Autumn Issue)
  6. Sergey Karelov, “Artificial Intelligence as A Geopolitical Factor”, 2018
  7. Stephan De Spiegeleire,  Matthijs Maas, Tim Sweijs, “Artificial Intelligence and The Future of Defense: Strategic Implications for Small- and Medium-sized  Force Providers”, The Hague Centre for Strategic Studies, 2017
  8. Stuart Russell, “Q&A: The Future of Artificial Intelligence,” University of California, Berkeley
  9. Tim Dutton, “An Overview of National AI Strategies”, Medium, 2018
  10. https://www.forbes.com/sites/bernardmarr/2018/02/14/the-key-definitions-of-artificial-intelligence-ai-that-explain-its-importance/#104a31884f5d