Authors: Oleg Shakirov and Evgeniya Drozhashchikh*
The creation of the Global Partnership on Artificial Intelligence (GPAI) reflects the growing interest of states in AI technologies. The initiative, which brings together 14 countries and the European Union, will help participants establish practical cooperation and formulate common approaches to the development and implementation of AI. At the same time, it is a symptom of the growing technological rivalry in the world, primarily between the United States and China. Russia’s ability to interact with the GPAI may be limited for political reasons, but, from a practical point of view, cooperation would help the country implement its national AI strategy.
The Global Partnership on Artificial Intelligence (GPAI) was officially launched on June 15, 2020, at the initiative of the G7 countries alongside Australia, India, Mexico, New Zealand, South Korea, Singapore, Slovenia and the European Union. According to the Joint Statement from the Founding Members, the GPAI is an “international and multistakeholder initiative to guide the responsible development and use of AI, grounded in human rights, inclusion, diversity, innovation, and economic growth.”
In order to achieve this goal, GPAI members will look to bridge the gap between theory and practice by supporting both research and applied activities in AI. Cooperation will take place in the form of working groups that will be made up of leading experts from industry, civil society and the public and private sectors and will also involve international organizations. There will be four working groups in total, with each group focusing on a specific AI issue: responsible AI; data governance; the future of work; and innovation and commercialization. In acknowledgment of the current situation around the world, the partners also included the issue of using AI to overcome the socioeconomic effects of the novel coronavirus pandemic in the GPAI agenda.
In terms of organization, the GPAI’s work will be supported by a Secretariat to be hosted by the Organisation for Economic Co-Operation and Development (OECD) and Centres of Expertise – one each in Montreal and Paris.
To better understand how this structure came to be, it is useful to look at the history of the GPAI itself. The idea was first put forward by France and Canada in June 2018, when, on the eve of the G7 Summit, Justin Trudeau and Emmanuel Macron announced the signing of the Canada–France Statement on Artificial Intelligence, which called for the creation of an international group to study AI-related issues. By that time, both countries had already adopted their own national AI development strategies – Canada was actually the first country in the world to do so in March 2017. The two countries proposed a mandate for the international group, then known as the International Panel on Artificial Intelligence, at the G7 conference on artificial intelligence in late 2018. A declaration on the creation of the group was then made in May 2019, following a meeting of the G7 Ministers responsible for digital issues. The group was expected to be formally launched three months later at the G7 Summit in Biarritz, with other interested countries (such as India and New Zealand) joining.
However, the initiative did not receive the support of the United States wfithin the G7. Donald Trump and Emmanuel Macron were expected to announce the launch of the group at the end of the event, but the American delegation blocked the move. According to Lynne Parker, Deputy Chief Technology Officer at the White House, the United States is concerned that the group would slow down the development of AI technology and believes that it would duplicate the OECD’s work in the area. The originators of the idea to create the group (which received the name Global Partnership on Artificial Intelligence in Biarritz) clearly took this latter point into account, announcing that the initiative would be developed under the auspices of the OECD.
A Principled Partnership
Like other international structures, the OECD has started to pay greater attention to artificial intelligence in recent years, with its most important achievement in this area being the adoption of the Recommendation of the Council on Artificial Intelligence. Unlike other sets of principles on AI, the OECD’s recommendations were supported by the governments of all member countries, as well as by Argentina, Brazil, Colombia, Costa Rica, Peru and Romania, which made it the first international document of its kind. They were also used as the basis for the Global Partnership on Artificial Intelligence.
In accordance with the OECD recommendations, signatory countries will adhere to the following principles of AI development: promote AI technologies for inclusive growth, sustainable development and well-being; the priority of human-centred values and fairness throughout the life-cycle of AI systems; the transparency and (maximum possible) explainability of AI algorithms; the robustness, security and safety of AI systems; and the accountability of AI actors.
In addition to this, the document proposes that the following factors be taken into account when drafting national AI development strategies: investing in AI research and development; fostering a digital ecosystem for AI research and the practical implementation of AI technologies (including the necessary infrastructure); shaping national policies that allow for a smooth transition from theory to practice; building human capacity and preparing for labour market transformation; and expanding international cooperation in AI.
A few weeks after the OECD endorsement, the recommendations on AI were included as an annex to the G20 Ministerial Statement on Trade and Digital Economy dated July 9, 2019, albeit with slightly different wording. The principles thus received the support of Russia, China and India.
Within the OECD itself, the recommendations served as an impetus for the creation of the OECD AI Policy Observatory (OECD.AI), a platform for collecting and analysing information about AI and building dialogue with governments and other stakeholders. The platform will also be used within the framework of the Global Partnership on Artificial Intelligence.
Artificial Intelligence and Realpolitik
The decision of the United States to join the GPAI was likely motivated more by political reasons than anything else. In the run-up of the G7 Science and Technology Ministers’ Meeting in late May 2020 (where all participants, including the United States, officially announced the launch of the GPAI), Chief Technology Officer of the United States Michael Kratsios published an article in which he stated that democratic countries should unite in the development of AI on the basis of fundamental rights and shared values, rather than abuse AI to control their populations, which is what authoritarian regimes such as China do. According to Kratsios, it is democratic principles that unite the members of the GPAI. At the same time, Kratsios argues that the new coalition will not be a standard-setting or policy-making body, that is, it will not be a regulator in the field of AI.
The United States Strategic Approach to the People’s Republic of China published in May 2020 and the many practical steps that the American side has taken in recent years are a reflection of the tech war currently being waged between the United States and China. For example, the United States has taken a similar approach to the formation of new coalitions in the context of 5G technologies. In 2018–2019, the United States actively pushed the narrative that the solutions offered by Huawei for the creation of fifth-generation communications networks were not secure and convinced its allies to not work with Beijing. Thirty-two countries supported the recommendations put forward at the Prague 5G Security Conference in May 2019 (the Prague Proposals), which included ideas spread by the United States during its campaign against Huawei (for example, concerns about third countries influencing equipment suppliers).
The United States is not the only GPAI member that is concerned about China. Speaking back in January about the U.S. doubts regarding the Franco–Canadian initiative, Minister for Digital Affairs of France Cédric O noted, “If you don’t want a Chinese model in western countries, for instance, to use AI to control your population, then you need to set up some rules that must be common.” India’s participation in the GPAI is particularly telling, as the United States has been trying to involve India in containing China in recent years. The new association has brought together all the participants in the Quadrilateral Security Dialogue (Australia, India, the United States and Japan), which has always been a source of concern for Beijing, thus sending a very clear signal to the Chinese leadership.
The Prospects for Russia
The political logic that guides the United States when it comes to participating in the Global Partnership on Artificial Intelligence may very well extend to Russia. The Trump administration formally declared the return of great power competition in its 2017 National Security Strategy. In Washington, Russia and China are often referred to as the main rivals of the United States, promoting anti-American values.
When assessing the possibility of interaction between Russia and the GPAI, we need to look further than the political positions of the participants. According to the Joint Statement from the Founding Members, the GPAI is open to working with other interested countries and partners. In this regard, the obvious points of intersection between Russia and the new association may produce favourable conditions for practical cooperation in the future.
First of all, the GPAI members and Moscow rely on the same principles of AI development. Russia indirectly adopted the OECD recommendations on artificial intelligence when it approved the inclusion of the majority of their provisions in the Annex to the G20 Ministerial Statement on Trade and Digital Economy in 2019 and thus shares a common intention to ensure the responsible and human-centred development and use of artificial intelligence technologies. This does not mean that there will not be differences of opinion of specific issues, but, as we have already noted, in its current form, the activities of the GPAI will not be aimed at unifying the approaches of the participants.
Second, according to media reports, Russia is working to re-establish ties with the OECD. It is already helping the OECD with its website, periodically providing data on new legal documents that will create a framework for the development and implementation of AI that have been adopted or are being considered.
Third, the current development of the national AI ecosystem in Russia shows that the state, business and the scientific community are interested in the same topics that are on GPAI agenda. This is reflected in the National Strategy for the Development of Artificial Intelligence for the Period up to the Year 2030 adopted in October 2019 and the draft Federal Project on the Development of Artificial Intelligence as Part of the National Programme “Digital Economy of the Russian Federation.” Furthermore, following the adoption of the National Strategy last year, Russian tech companies set up an alliance for AI development in conjunction with the Russian Direct Investment Fund, which is very much in keeping with the multistakeholder approach adopted by the Global Partnership on Artificial Intelligence.
It would seem that politics is the main stumbling block when it comes to Russia’s possible participation in GPAI initiatives, for example, the organization’s clear anti-Chinese leaning or its members openly discrediting Russia’s approaches to the development of AI. That said, Russia has nothing to gain from politicizing the GPAI, since cooperation with the organization could help it achieve its own goals in artificial intelligence. What is more, we cannot rule out the possibility that the GPAI will be responsible in the future for developing unified AI rules and standards. It is in Russia’s interests to have its voice heard in this process to ensure that these standards do not turn into yet another dividing line.
*Evgeniya Drozhashchikh, Ph.D. Student in the Faculty of World Politics at Lomonosov Moscow State University, RIAC Expert
From our partner RIAC