In July 2017 the Belfort Center for Science and International Affairs at the Harvard Kennedy School published the report Artificial Intelligence and National Security, arguing that in the future artificial intelligence could become a transformative medium comparable to nuclear weapons, aircraft, computers, and biotech security technology. It is therefore reasonable to include Artificial Intelligence in discussions that may influence international relations.
The international order includes two basic aspects, one being the power structure and balance of power of the major countries and groups of States, and the other being the rules that should be followed in the management of relations between these countries.
The international order is a state of affairs in which countries in the global system should adopt non-violent methods to resolve conflicts in accordance with international rules. Its constituent elements are jus gentium, shared rules and regulations, and relevant institutions.
Changes in this system are essentially caused by changes in the international structure, but the structure is not a constitutive element of the international order. Hence, if the aforementioned fallacious system is to be changed, a new international order must be established, which is nothing more than the redistribution of power, i.e. the core content of the rearrangement of international institutions.
Artificial Intelligence can have an impact on existing international rules and give rise to a new jus gentium by changing the balance of power and mutual relations of international players, thus influencing changes.
First of all, Artificial Intelligence will economically affect the balance of power between countries and even trigger a new cycle of influence and co-management by the great powers.
As early as 1989 Paul Kennedy argued – in his book The Rise and Fall of Great Powers. Economic Change and Military Conflict from 1500 to 2000 – that in the long run there was an obvious link between the economic rise and fall of every great world power. In June 2017 Pricewaterhouse Coopers published Seize the Opportunity. 2017 Summer Davos Forum Report predicting that by 2030 the AI contribution to the world economy would reach 15.7 trillion US dollars and that the People’s Republic of China and North America were expected to become the largest beneficiaries, totalling 10.7 trillion US dollars.
In September 2018 the report Frontier Notes: Using Models to Analyse the Impact of Artificial Intelligence on the World Economy, published by the McKinsey Global Institute, estimated that Artificial Intelligence would significantly improve overall global productivity. Excluding the impact of competition and transformation cost factors, Artificial Intelligence could contribute an additional 13 trillion US dollars to global GDP growth by 2030, with an average annual GDP growth of around 1.2 per cent.
This is comparable to or greater than the transformative impact of many other technologies throughout history, such as steam engine in the 19th century, industrial production in the 20th century, and information technology in the current century. The report also pointed out that countries and regions (mainly developed economies) with leading positions in Artificial Intelligence can achieve economic growth of 20 to 25 per cent on current basis, while emerging economies may only record half this rate.
The Artificial Intelligence gap can lead to further deepening the digital divide. Artificial intelligence can change the global industrial chain. The new industrialisation represented by industrial robots and smart manufacturing will lure the manufacturing industry back to developed economies, and will impact the de-industrialisation of many developing countries sooner than expected. Hence opportunities would remain locked in the country that provides “only” the resource or raw material.
The development and implementation of Artificial Intelligence require a lot of funding, high-tech content, and can lead to changes in the employment structure, making highly repetitive and low-tech jobs gradually disappear.
Moreover, in another McKinsey report from 2017, based on research into 46 countries, it was predicted that by 2030 as many as 800 million people worldwide would lose their jobs and be replaced by automated robots. There will be a massive displacement of jobs worldwide similar to that seen in the early 20th century, when most of the world’s jobs moved from agriculture to industry. At the same time, the widespread implementation of AI technology will also increase the demand for professionals in this field.
According to the research, there are three types of countries that are most likely to benefit from the development of AI technology.
The first type consists of countries with first-rate advantages in Artificial Intelligence – such as the United States and China – and they are all favoured.
The second type is represented by capital- and technology-intensive countries with a small population or a downward trend, such as Japan, South Korea and Singapore, which not only have the capital and technical conditions to develop Artificial Intelligence, but can also use the AI development to compensate for a lack of total population or a downward trend, as well as an ageing population structure and other disadvantages.
The third type includes countries with more scientists, mathematicians, engineers or States that value and appreciate vocational training related to science, technology, engineering and mathematics (STEM).
Over and above the economic issue, Artificial Intelligence will militarily change the balance of power between countries. Proponents of military AI believe it will subvert the form and style of warfare. Armoured or mechanized warfare uses materials to release energy, relying on oil and steel, whereas computerised warfare uses networks to gather energy, relying on information, links and connections. According to current expectations, once warfare enters the AI era, it will be a confrontation of robots and automation, controlled by the aforementioned AI.
It can be expected that, under AI conditions, elements of warfare such as combatants, battle concepts and winning mechanisms will change completely. In a traditional war, even if there is a gap in terms of weapons and training levels between the opposing sides, the disadvantaged side can anyway fight with favourable times and places, superior strategies and advanced tactics. For example, during the wars in Iraq and Afghanistan, improvised explosive devices (IED) caused problems for the US and Soviet armies in the latter country. Under the conditions of intelligent warfare, instead, the technological contribution of one side through AI will quickly create an overwhelming advantage on the battlefield, thus making it impossible for the weaker side to define an effective observation-judgement-decision-action cycle, always remaining in a passive position.
In its report Artificial Intelligence Changing the World, the Brookings Institution put forward the concept of “hyperwar”, i.e. war is a process of race against time and usually the side with the fastest decision-making and execution process will prevail. The decision-making speed of the AI-assisted command and control system will far exceed that of the traditional mode of warfare – combined with the automatic weapon system that can autonomously decide to launch lethal weapons – and will greatly accelerate the warfare process, so that a new expression – “extreme speed warfare” – will and must be coined to describe this mode of warfare. Regarding the latter, the article What Happens When Your Bomb-Defusing Robot Becomes a Weapon, published by Caroline Lester on the website The Atlantic on 26 April 2018, used many analyses to show that military robots can significantly reduce the threat of roadside bombs, with all due respect to Iraqi and Afghan patriots.
Artificial Intelligence will also lead to revolutionary changes in military equipment. Unmanned lethal automatic weapon cluster combat could become the protagonist and the main combat method in future wars.
Once the aerial drone, the unmanned submarine, the ground robot, the unmanned tank, the war of attrition and sea tactics are perfected, they will make large-scale weapon platforms – such as aircraft carriers and fighter aircrafts – complex and expensive, as the latter are less advantageous from the viewpoint of warfare cost and combat effectiveness.
It is as if an F-35 fighter jet, with a single cost of hundreds of millions of dollars and with one man on board fighting a group of low-priced armed drones, is equivalent to shooting mosquitoes.
What needs to be explained is that there is still great uncertainty about the AI impact on armies: it is not known to what extent and how this impact will manifest itself. In the discussion held at the Conference Artificial Intelligence and Security of the 7th World Peace Forum at Tsinghua University in July 2018, some experts pointed out that although the future development trends of machine learning in industrial robots, material science and other technologies can be generally studied, the specific impact of the mix of these technologies on future warfare cannot be accurately assessed. In the first thirty years of the 20th century, European military powers such as Germany, Great Britain, the Soviet Union, France and Italy developed tank, aircraft, missile and radio communication technologies. Nevertheless, it was only after Germany had waged the Blitzkrieg in World War II that the world discovered that those new technologies together could bring such unimaginable changes to warfare. Now, regardless of algorithmic warfare or similar tactics, the heated debate in strategic circles is still to analyse the AI impact on the operations of a single technology. Without a holistic understanding of the military applications of AI technology, the planned countermeasures could become a costly and unnecessary new Maginot Line. (2. continued).