- 8 -
Artificial Intelligence And Global Geopolitics

June 29, 2016

Artificial Intelligence And Global Geopolitics – From "Atoms for Peace" to "AI for Mankind".

 

Artificial Intelligence (AI), a top priority for the ubiquitous American tech companies, for Industry 4.0 or digital China, is already reshaping global business, but this major scientific and technological disruption will also deeply impact the relations between powers.

While narrow AI has moved from the labs to our daily lives, informed personalities like Stephen Hawking, Nick Bostrom, Bill Gates or Elon Musk have rightly raised concerns about the risks inherent to a strong AI capable of equaling or even surpassing human intelligence.

Anticipating the emergence of an even more powerful and increasingly autonomous AI reinforced by quantum computing, these engaged voices are asking for a collective reflection upon what could constitute an external challenge to mankind, a technology which could dominate its creator.

The recent win of the AlphaGo computer program over the Korean go champion Lee Sedol was indeed a strong signal of the rapid development of machine learning at the intersection of computer science and neuroscience.

However, a more immediate danger connected with the advancement of intelligent machines is an AI fracture enlarging what is already known as the digital divide. While AI's algorithms and big data increase the productivity of a small segment of the global village, half of the world population still does not have access to Internet. "Don't be evil" can be Google's slogan, but exponential technologies carry with them the risks of unprecedented inequalities.

While AI's social and political effects are often discussed, the geopolitical implications of the "Fourth Industrial Revolution" have been surprisingly absent from the public debates.

How AI could affect the Sino-Western relations or, more specifically, the Sino-American relations, the major determinant of today's international order? For decades, nuclear weapons stood as the frightening symbols of the Cold War. Will AI become the mark of a 21st century Sino-Western strategic antagonism?

For humanity, the atomic age has been a time of paradoxes. In the aftermath of the 1945 Hiroshima and Nagasaki nuclear bombings, an arms race involving the most lethal weapons defined the US-Soviet relations in what also constituted a permanent existential threat to human civilization. But, analysts will also argue that it is the Mutually Assured Destruction (MAD) doctrine acting as a deterrent among rational actors which prevented a direct conflict between the two superpowers.

As the 2015 Plan of Action for Iran's nuclear program demonstrates, 70 years after Hiroshima and Nagasaki, world powers actively collaborate to avoid nuclear proliferation, even if North Korea appears to be a counter example of this dominant trend.

But the Sino-Western convergence of views on the issue of nuclear proliferation does not apply in the cyberspace. Despite a certain level of interconnection between some private Chinese and American Internet companies and financial institutions, the overall Sino-American relations in the cyberspace are characterized by strategic mistrust.

Besides, in space science and in the exploration of the universe, the US and China are unfortunately following two separate courses. While China prepares to operate her own modular space station, the International Space Station (ISS) shows that in this strategic field the West can work with Russia but that Sino-Western synergies are almost impossible to reach.

Any responsible approach to AI has to take into account the combined lessons of the atomic age, of the digital dynamics and of the space exploration. Should Western AI and Chinese AI develop on two separate trajectories, one would dangerously increase the risks of creating an irreversible Sino-Western strategic fracture, for AI does not increase power in a limited quantitative manner but it modifies its nature.

In this context and following the appreciation of the interactions between AI and global politics, an "International Artificial Intelligence Agency" should be established inspired by the International Atomic Energy Agency (IAEA).

It is in the "Atoms for Peace" address to the United Nations General Assembly that US President Dwight D. Eisenhower (1890–1969) proposed in 1953 the creation of the IAEA. Today, our actions must be guided by the spirit of "AI for Mankind".

A United Nations International Artificial Intelligence Agency involving academics, private businesses, the world civil society and, of course, the governments should at least give itself the following four objectives.

First, it has to create the conditions for AI's awareness across our societies and for a debate to take place on AI's ethical implications. Scientists, engineers, entrepreneurs, legal experts, philosophers, economists have to analyze AI from all possible angles, its future(s), its potential effects for humanity.

Second, this international body should take all possible actions to prevent an AI fracture which would dangerously enlarge the digital divide. One can't accept to have, on one side, a tiny segment of humanity making use of a series of Human Enhancement Technologies (HET) and, on the other side, the vast majority of the world population becoming de facto diminished, what transhumanism revealingly abbreviates as H+ can't be a plus for a few and a minus for all the others.

Third, the agency should ask for transparency in the AI research at both the governmental and the corporate level. The issue of nuclear proliferation and therefore the creation of the International Atomic Energy Agency (IAEA) followed the secretive Manhattan Project and the use of nuclear bombs to end the war in the Pacific, if humanity really wants to protect itself from the military use of strong AI and its tragic consequences it has to define a set of rules and policies which would maintain research within reasonable and collectively accepted limits. The IAEA imperfectly manages an existing threat, the AI agency would aim at preventing the realization of what could be an even greater danger.

Fourth, an international AI body should encourage knowledge sharing and international cooperation. Elon Musk's OpenAI initiative is certainly a constructive force encouraging openness and collaboration but the "AI for Mankind" ideal can not depend only on a group of private entrepreneurs.

Artificial Intelligence, more than any other technology, will impact the future of mankind, it has to be wisely approached on a quest toward human dignity and not blindly worshiped as the new master of a diminished humanity, it has to be a catalyst for more global solidarity and not a tyrannical matrix of new political or geopolitical divisions.