By Amruta Mahuli1, Sumit Kumar2 and Suryesh K Namdeo1
[1] DST Centre for Policy Research, Indian Institute of Science,
[2] Monash Sustainable Development Institute, Monash University
Illustration by Khyati Trehan, selected by Vivian Zhao
Edited by Merissa Hickman
Abstract
Artificial intelligence (AI) has emerged as one of the defining technologies of our time. Countries worldwide are developing their AI strategies to develop their national capacities, increase economic competitiveness and growth, and address local, regional, and global issues. The vast implications of AI are guiding foreign policy actions of different countries as they find AI as a new avenue for international cooperation and competition. In this study, we review the emerging contours of AI strategies and partnerships at national, bilateral, and multilateral levels by analysing the approach of some major countries and international forums. Here, we discuss the national strategies, bilateral collaborations, and structure and function of relevant international platforms to understand the evolving nature of international partnerships on AI. Further, we highlight the disconnect between global priorities and national strategies, collaborations among some developed and developing countries, and the ongoing deliberation on global norms and standards. Finally, we put forward some considerations for global cooperation to develop rules, norms and standards for AI.
Science to Policy Statement
Building capabilities in AI is becoming increasingly important among the major economies of the world because of its economic and military implications. Moreover, AI governance and regulation in a country are being impacted by the larger geopolitical discourse. Thus, it is essential to understand the importance of AI in international relationships to develop it as a technology that promotes cooperation instead of competition.
Key Words: artificial intelligence, international relations, international partnerships, global AI policies
Introduction
With a large number of people, corporations, and government entities using the internet for day-to-day communications and transactions, it is argued that digital technology is becoming a driving geopolitical force shaping international politics [1]. This is particularly evident from the increase in the adoption of emerging digital technologies like AI in the different foreign ministry operations (or their foreign equivalents, such as a Department of State) through tools like chatbots for consular services, use of image recognition and sentiment analysis for gauging public response, and virtual
diplomatic missions [2,3].
In general, various national policies on artificial intelligence (AI) currently aim to promote the use of AI for economic growth while putting in place rules for data protection, data localization and privacy. These policies also protect data that is on servers that may become vulnerable to cyberattack and unauthorised access to data, which could be unlawfully used for data theft or data corruption by malicious third parties actors [4,5]. Additionally, AI policy aims to codify the control, ownership and beneficiaries of the result of data use and analysis as well as the development of ‘trustworthy’ and ‘responsible’ AI by software developers [6,7]. Therefore, analysing such adopted digital diplomacy practices is especially critical with the growing trend of digital trades and transactions across countries through crossborder data flow, which is estimated to increase by 29% in bandwidth by 2022 as compared to 2021. The AI economy, too, reflects a similar estimated increase globally from $35.92 billion to $47.47 billion with AI applications like self-driving cars and facial recognition technology products expecting a growth trend from 22.7% and 15.7% annual CAGR growth, respectively, during the forecast periods of 2021-2026 and 2021-2028 [8].
Policies are national, but data flows beyond their geographic and political silos across the internet ocean. With a few exceptions like the European Union’s General Data Protection Regulation (GDPR) compliance adherence, international law has lagged behind the development of rapidly emerging technologies, like AI. Given the increasing importance of emerging technologies like AI, it is argued that the nation-states have the option to either ensure greater geopolitical cooperation or geopolitical competition [9,10]. National policies, therefore, currently dominate AI regulatory rules (or lack thereof), and are vying for alliances to gain a majority consensus for international regulation for AI.
Research Methodology
This paper grounds itself in the assumption that a competitive, and desultory disorganised approach to AI in diplomacy may prove to be short-sighted, particularly as AI is increasingly being adopted as a diplomatic tool to retain or redistribute power on a global scale. Therefore, this exploratory paper accepts the arguments made in the extant digital diplomacy discourse that encourages the states to take concrete, strategic steps towards international cooperation and applications of emerging technologies, particularly in AI [11-13]. The methodology employed in this article involves triangulating the patterns in examination of international initiatives and platforms, national AI strategy documents, and bilateral agreements , and social media engagement to unravel patterns related to the importance granted to AI in existing foreign policy objectives and practices. The conclusion summarises the analysis of emergent patterns that provide insights on diverse international cooperation and competition strategies with respect to AI.
The scope of this study is limited to one country each selected to represent four continents, with the exception of Asia, which has two representative nations: South America (Brazil), Asia (China and India), Europe (Germany), and North America (the United States). These countries were chosen for their diverse geographic and socioeconomic demographics reflected in their income profiles, GDP and technological innovation levels [14-16]. Two countries from Asia, China and India were chosen for this study as they have the largest populations and have been investing heavily in AI with possible global implications [17]. The countries under study have varied Global AI index rankings [India-17, China-2, Brazil-39, Germany-9 and the US-1]. The Global AI Index identifies a nation’s capacity for investment, innovation, and implementation of AI [18]. They have all also formulated their national strategy roadmaps to use AI for economic growth [19]. Additionally, these countries are also part of important international alliances such as the Global Partnership for Artificial Intelligence (GPAI) and the Quadrilateral Security Dialogues (QUAD). Some of them are parties to frameworks such as UNESCO agreement on the Ethics of AI, G20 AI principles, OECD AI Principles thus building their capabilities and using these alliances for competitive advantage against other economies. This, coupled with these countries attracting the highest number of patents in the field of AI, with China being the world leader, makes them good candidates for this study [20].
To carve out patterns, this paper uses a non-experimental mixed methodology to collect, triangulate, and analyse secondary data [21]. For qualitative analysis, the preliminary research method for this study is a literature review and conceptual modelling of the existing discourse around digital diplomacy, state practices around AI governance and regulation, and the various multilateral and bilateral collaborations between nation-states that specifically refer to AI or AI based technologies (referred by AI-related agreements in this paper and details of which are provided in Appendix 1). It draws data from national and international AI strategy documents and treaties, agreements, and AI specific international coalitions for qualitative analysis where the term ‘contracting countries’ refers to the countries having formal partnership or agreements in the area of AI.
For data visualisation and data representation of bilateral agreements, Datawrapper, a third-party software, was used [22]. Using global symbol maps available on Datawrapper (V2022) and the data in Appendix 2, the visual representations using maps highlighting the contracting countries have been showcased in the third section of this paper. This section in the paper also showcases histograms that can aid in the analysis of the bilateral agreements and the demographics of contracting parties in these agreements to draw out any existing patterns using parameters like the contracting party’s per capita income (GDP nominal) ranking as of March 2022 and their ranking in the Global Innovation Index (GII) 2021 [23,24], data for which can be found in Appendix 2.
Considering that diplomacy for AI and AI in diplomacy is still in a nascent stage across the globe, it can be assumed that scale and capacity of innovation may also be a matter of consideration for increased engagement. This study, therefore, relies on WIPO’s GII to gauge any patterns in the contracting countries. These indices have been particularly chosen to understand any underlying patterns that the countries under this study use to select their contracting parties in these bilateral agreements.
National Strategies and Bilateral Cooperation on AI
In this section, we look at the national strategies or policies and bilateral agreements of the countries under study to unravel major highlights and trends related to international cooperation on AI that are helping to form or strengthen political and economic alliances.
India. India’s two-part national strategy document on AI, National Strategy for Artificial Intelligence #AIforAll, relies on seven broad principles: safety and reliability, equality, inclusivity and non-discrimination, privacy and security, transparency, accountability, and protection and reinforcement of positive human values [25-27]. Keeping the much-needed collaborative spirit for adoption and application of AI in mind, the second part of the Indian strategy focuses on coordination and collaboration with state/non-state actors within and outside India. It recognizes the government’s role in catalysing partnerships, access to infrastructure, and fostering innovation amongst various private and public institutions. It envisions an umbrella organisation, the Centre for Studies on Technological Sustainability (CSTS), that will study the AI landscape and foster globally competitive technological development.
1a.
1b.
1c.
Figure 1: a) Visual representation of countries contracting with India for AIrelated agreements. (Created with Datawrapper.) b) Histogram representing the frequency for per capita GDP ranking of India’s contracting countries (in nominal terms). c) Histogram representing the frequency for GII ranking of India’s contracting countries.
Almost 60% of India’s contracting countries are from the European continent, which may be due to Europe’s strong position in innovation of emerging technologies. This is followed closely by engagements and collaborations with Asian countries like Indonesia, Uzbekistan and Myanmar in addition to Australia and Brazil (Figure 1a). It is also worth noting that the economic profile of these contracting countries skews more towards countries with higher per capita income apart from Indonesia, Uzbekistan and Myanmar (Figure 1b). As can be observed from Figure 1c, the GII ranking of contracting countries skews towards the left as India engages and collaborates with countries that rank higher on the GII with the likes of Switzerland which ranks 1st in GII out of 132 economies considering 81 different indicators for innovation tracking. This shows India’s willingness to cooperate with countries that are technologically advanced while also establishing linkages with some strategically important countries in the Global South. This approach is consistent with the paradigm of ‘strategic autonomy’ India applies in its foreign policy [28]. A substantial majority of the agreements that are publicly available on the official foreign ministry portal are generic agreements for science and technology cooperation with a push for Information and Communication Technologies (ICTs). Only agreements with Canada, Finland, Brazil, Australia and Portugal are specific contracts – focused on distinct aspects of AI that are detailed in Appendix 1.
Overall, India’s national strategy on AI focuses on principles such as safety, inclusivity, transparency, while its primary partnerships are with technologically advanced high-income countries.
China. China’s ‘Next Generation Artificial Intelligence Development Plan’ acknowledges China’s strong foundations in AI and seeks to advance its development and deployment across different sectors [29]. Even though the plan contains several specific recommendations, details regarding international cooperation or competition have been largely left out [30,31]. Notwithstanding this assessment, it must be noted that the plan provides for facilitation of indigenous AI enterprises to cooperate with international research institutes and teams to comprehensively plan their innovation resources using international linkages. It highlights leveraging AI-based technologies that provide a competitive advantage internationally. It aims to integrate international cooperation on AI in its ‘One belt, One Road’ strategy (now the Belt and Road Initiative). It also calls for promoting the establishment of international AI organisations, of which it would be a part, and jointly formulating related international standards [30].
2a.
2b.
2c.
Figure 2: a) Visual representation of countries contracting with China for AI-related agreements. (Created with Datawrapper.) b) Histogram representing the frequency for per capita ranking of China’s contracting countries. c) Histogram representing the frequency for GII rankings of China’s contracting countries.
Almost 80% of China’s contracting countries are from the European continent, followed by engagements and collaborations with Asian countries and some Latin American countries (Figure 2a). It is also worth noting that the economic profile of these contracting countries skews more towards countries with higher per capita income (Figure 2b). Considering that China itself ranks 14th for its innovative capabilities, its engagement with countries that rank lower than itself is interesting and is probably part of China’s larger strategic calculations (Figure 2c) [32,33]. The competitive tensions between the Global North and China may also be a contributing factor to its choice of contracting parties. Substantially, majority of the agreements that are publicly available on the official foreign ministry portal are generic agreements for science and technology cooperation with a push for ICTs with only agreements with Russia, Norway and Germany having specific contracts that are focused on distinct aspects of AI, described in Appendix 1b.
In summary, China’s national strategy aims to advance the development and deployment of AI across sectors, while facilitating cooperation between indigenous AI enterprises and international research institutes. Here, a significant majority of agreements are generic science and technology agreements with high income countries.
Brazil. In 2021, the Brazilian government adopted a strategy on AI for the Brazilian state to: a) contribute to the development of ethical principles for responsible AI, b) promote sustained investment in AI research and development, c) remove barriers to AI innovation, d) empower and train professionals for the AI ecosystem, e) stimulate innovation and the development of Brazilian AI in an international environment, and f) promote an environment of cooperation between public and private entities like the Ministry of Science, Technology and Innovation, the Brazilian Industry Research and Innovation Company, industry, and 17 research centres for the development of AI.
Brazil’s strategy for international collaboration and cooperation in AI hopes its national network will ‘intensify international exchange of knowledge and reciprocal collaboration’ with countries in Europe, Israel, UK and North America but lacks further details explaining execution of the plan [34-36].
3a.
3b.
3c.
Figure 3: a) Visual representation of countries contracting with Brazil for AIrelated agreements. (Created with Datawrapper.) b) Histogram representing the frequency for per capita ranking of Brazil’s contracting countries. c) Histogram representing the frequency for GII rankings of Brazil’s contracting countries.
Like India and China, most of Brazil’s contracting countries (almost 60%) are from the European continent, followed by 15% of its engagements and collaborations with Asian countries (India and China) and North America (US and Canada) along with Australia (Figure 3a). It is also worth noting that the economic profile of these contracting countries is highly skewed towards countries with higher per capita income and global innovation indices ranging from 1 to 46, reflecting Brazil’s ambition to employ know-how from the world’s leading countries in AI innovation and research (Figures 3b & 3c). The majority of publicly available agreements on the official foreign ministry portal are generic. A push for ICTs having specific contracts that are focused on distinct aspects of AI exists only with India, China, US, Italy, and Australia, as detailed in Appendix 1.
Overall, the Brazilian strategy focuses on promoting innovation in AI through investment in research and development. It has established international cooperation mostly with high income and more innovative countries.
Germany. In 2018, the German Federal Government launched its Artificial Intelligence Strategy aimed at making Germany and the EU global leaders in the development and use of AI technologies, safeguarding the responsible development and use of AI, and integrating AI in society in ethical, legal, cultural and institutional terms (The Federal Government 2020). However, the COVID-19 global pandemic induced geopolitical and sociocultural shifts that prompted updates to its AI strategy in 2020. These shifts prioritised the technical expertise, research, transfer and application, regulatory framework and society–with a strong emphasis on issues of pandemic control, environmental and climate protection, and European and international networking—in an effort to foster the spirit of ‘internationalisation of AI research’ [37].
The federal government’s international cooperation and collaborative efforts can be grouped into four categories: a) its efforts within the EU to bolster the network and mobilise expertise across EU to establish ethical guidelines and develop basic and applied research in the medium and long term; b) its efforts for cooperation with France, Canada, and Japan to promote collaborative research and development projects in the context of ‘Industry 4.0’, c) its efforts to establish contact in Silicon Valley as a ‘networking centre’, and d) its efforts towards Global South partner governments to create global public goods and implementation capacities in the field of AI. Additionally, the AI strategy plan also aspires to establish a network of think tanks focused on international standardisation and regulatory forums as a part of ‘international AI governance’ to further the pursuit of ethical and equitable AI in the world.
4a.
4b.
4c.
Figure 4: a) Visual representation of countries contracting with Germany for AI-related agreements. (Created with Datawrapper.) b) Histogram representing the frequency for per capita ranking of Germany’s contracting countries. c) Histogram representing the frequency for GII rankings of Germany’s contracting countries.
Of all the countries in this study, Germany’s engagement for international AI cooperation is the most diverse, with approximately 36% of its contracting countries being from the European continent, followed by engagements and collaborations with Asian (India, China, South Korea, Singapore and Japan), North American (US, Canada and Mexico), South American (Brazil and Columbia), African countries (South Africa), and Australia (Figure 4a). It is also worth noting that the economic profile of these contracting countries skews more towards countries with higher per capita income (Figure 4b). Global innovation indices of contracting countries ranging from 3-68 with the majority of contracting countries falling within 3-30 in their ranking in the global innovation indices, reflecting Germany’s ambition to employ know-how from the world’s leading countries in the field of AI innovation and research (Figure 4c). Most publicly available agreements (~57%) on the official foreign ministry portal are generic agreements, except a push for ICTs with Brazil (almost 42%), as detailed in Appendix 1d.
In summary, Germany has a diverse international engagement on AI and has a strategy that aims to make it a global leader while safeguarding the responsible development and use of AI and integrating it into society.
United States. The 2019 Executive Order 13859 enlisted several projects for the National Artificial Intelligence initiative to promote and ensure US leadership in the space of development and innovation of AI [38]. In addition to ensuring AI infrastructural needs are met, it prioritises effective methods for AI collaborations within the federal government. The following year’s annual report endorses – supporting international collaborations and partnerships in line with its national values, those ‘grounded in evidence-based approaches, analytical research and multistakeholder engagement’ [39]. The report further provides evidence of its international participatory engagement over the past years: the G7 Innovation and Technology Ministerial, G20 Digital Economy Ministerial, NATO, the European Union, OECD, and such other multilateral discussions and bilateral agreements with its allies on matters of mutual agreement.
5a.
5b.
5c.
Figure 5: a) Visual representation of countries contracting with the United States for AI-related agreements. (Created with Datawrapper.) b) Histogram representing the frequency for per capita ranking of the US’s contracting countries. c) Histogram representing the frequency for GII rankings of the US’s contracting countries.
At the outset, it must be noted that compared to other countries studied here, the number of publicly available contracts specifically relating to the US is smaller than other countries (Figure 5a). However, the contracting countries are more diverse than countries that have been studied so far in this section. Despite its diversity, like other countries so far, almost 40% of US’s contracting countries are from the European continent while the rest belong to North America (Canada), Asia (Japan) and South America (Brazil). It is interesting to note that many close trade partners such as Mexico (a member of the North American Free Trade Agreement) and close allies such as most members of NATO are missing from this list. It is also worth noting that the economic profile of these contracting countries skews more towards countries with higher per capita income, just like the global innovation index ranking (Figures 5b & 5c). Unlike the countries studied so far, many agreements that are publicly available on the official Department of State portal are specific agreements that focus on distinct aspects of AI, with a detailed in supplementary material 1e. Only its agreement with Brazil is a generic science and technology cooperation with a push for ICTs.
Taken together, the US executive order highlights the efforts in place and process to ensure the US leadership in the area of AI by synergising efforts within the government and developing international partnerships. The US has a small number of focused international cooperation agreements on AI with a diverse set of countries.
Major International Initiatives and Discussion Platforms on AI
Next, we looked at the major multilateral and global platforms used for AI related policies and diplomatic negotiations to examine their current structure, objectives, and functionality. We find that AI has been an active area of discussion in several major international forums. A concise summary of major AI-related forums are below.
Global Partnership on Artificial Intelligence (GPAI). Launched in 2020, GPAI works as a multistakeholder initiative bringing together experts from science, government, industry, civil society and international organisations to promote multidisciplinary research and international collaborations on AI-related priorities [40]. It currently has 25 nation-states as members, and its thematic focus includes responsible AI, the future of work, data governance, – innovation, and commercialisation. GPAI is a product of G7 discussions and aims to implement some of the OECD recommendations on AI [41]. With many powerful democratic countries being its member, GPAI is likely to emerge as one of the most important global forums for discussions on setting up the norms and regulations for the global governance of AI. Due to its nature as a multi-stakeholder platform, it is arguably well placed to develop a global framework for AI cooperation and governance with inputs from all stakeholders. India, Germany and the US are among the founding members of GPAI.
OECD AI Policy Observatory. OECD, with its 37 member states, adopted the AI principles in May 2019 to promote use of AI which is innovative and trustworthy and that respects human rights and democratic values [41]. International cooperation for trustworthy AI is one of the key principles among several other principles adopted by OECD. This principle emphasises the role of the government to cooperate across borders and sectors to share information, develop standards and work towards development of responsible AI. The principle emphasises that sharing of knowledge is important in order to build long term expertise on AI, and develop technical standards which are interoperable and trustworthy. Developing common metrics to assess the performance of AI systems such as accuracy, efficiency, robustness and advancement of societal goals is also key for strengthening international cooperation in the field of AI [42]. OECD also functions as a secretariat to GPAI to facilitate synergy between GPAI’s technical work and the international policy leadership provided by the OECD.
UNESCO agreement on the ethics of AI. Adopted in 2021, UNESCO’s standards on the ethics of AI are the first truly global framework on AI [43]. It focuses on the use of AI for promoting the human rights and achievement of the SDGs and addresses the issues around accountability, transparency and privacy [44]. This agreement puts forward common principles and values to assist in the creation of the policy and governance infrastructure required to ensure the healthy development of AI. Even though the agreement is voluntary and non-binding, it signifies the consensus among the member states to work together for further developing a global governance framework on AI in the future. Interestingly, the US is not signatory of the agreement as it is not a UNESCO member.
The G7. The G7 countries which comprise Canada, France, Germany, Italy, Japan, the UK, and the US have been focusing on cooperation among the member countries in the field of AI given its important impact across all sectors and because of its strategic importance. AI has been one of the key agenda items of the G7 ministerial meetings since 2017. At the 2017 G7 ICT and Industry Ministerial meeting it issued a document titled ‘Multistakeholder Exchange on Human Centric AI for our Societies’ that highlighted the use of AI for solving the global challenges related to environment, transportation, or health [45]. In the G7 Innovation Ministerial meeting held in 2018, the countries highlighted the need for interconnectedness for innovation in AI, trust and inclusivity in the development or adoption of AI [46]. In the 2019 G7 Summit it released the ‘Strategy for an Open, Free and Secure Digital Transformation’ highlighting that AI is bringing radical transformation in the societies, economies and future of work, and AI could play an important role in achieving the Sustainable Development Goal by 2030 [47]. At the 2020 G7 Science and Technology Ministerial Meeting, it issued a Minister’s Declaration on COVID-19 in which it committed to the responsible and humancentric development and use of AI [48].
The G20. AI has become one of the key subjects of discussion among the member countries of the G20 nations as well. During the G20 meetings held in Osaka under Japan’s presidency in 2019, the member countries endorsed the G20 AI principles in line with OECD Principles on AI in the leader’s Declaration [49]. The declaration highlighted the need for responsible development and use of AI to help advance SDGs. The declaration also highlighted the need for international cooperation with developing countries and other stakeholders to advance the G20 Principles of AI including sharing of knowledge with other countries, and development of global standards and metrics to measure for building trustworthy AI systems [50].
Quadrilateral Security Dialogue (Quad). The Quad, a multilateral setting which comprises the United States, India, Japan and Australia, has set up a ‘Working Group on Critical and Emerging Technologies’ to ‘facilitate cooperation and innovative technologies of the future’ [51]. Further, a contact group on Artificial Intelligence and Advanced Communication will be established which would focus on ‘standard-development activities as well as foundational pre-standardisation research’ [52]. Another important initiative in this direction is the establishment of the Quad Tech Network that would promote track 2 research and public dialogue on critical technologies relevant to the Indo-pacific region [53]. This network is likely to have a focus on enhancing policy cooperation in AI among the Quad nations. Collectively, these are early indicators for things to come in terms of intense future cooperation on AI among Quad members.
Discussion
Given that all countries under this study developed some variation of a strategy or policy document on AI with an emphasis on capacity building, innovation, or international collaborations, it was observed that terms like ‘ethical’, ‘reliable’, ‘accountable’, ‘explainable’, ‘responsible’, ‘transparent’, and ‘trustworthy’ are emphasised in these national strategies and international forums. A number of emerging AI alliances were also mentioned in passing as to how they may figure into the participatory AI agreements. However, precisely how the purposes, objectives, interpretations, and manifestation of these buzzwords translate into reality (that is, if they lead to improved international cooperation or if they are merely guises for continued strategic competition) has yet to be determined. Therefore, there is a need for further analysis of these AI alliances in relation to their societal, economic, legal, and geopolitical implications.
Upon analysing the bilateral agreements signed by the subject countries, it was observed that the number and nature of international agreements vary widely among the countries studied. The US signed fewer agreements but with more specificity, while India, China and Brazil signed more agreements that were broader in scope, as detailed in Appendix 1. Among the countries studied, it was observed that emerging economies like India and China have the highest number of collaborations with countries in the Global North, while the US had the least number of collaborations, they were mostly with developed economies (Figures 1-5). It was also noted that a vast majority of bilateral agreements (signed by both developed countries like US, Germany and China and developing countries like Brazil and India) were signed with countries having a high GII ranking and high GDP per capita income, while there was little SouthSouth cooperation in the developing countries.
The primary explanation explored here is that these countries are still in the early stages of developing their AI ecosystem and therefore prefer linkages with advanced nations to expand their expert-base and seek co-development of their initial projects. However, the limited international collaboration among the Global South countries could also be an indication of their limited capacity or priority to invest in AI despite the need [54,55]. Forums like GPAI, G20 and OECD can play a big role in facilitating and empowering such capacitydeficient nations by enhancing their capacities and capabilities in AI. Given the nascent AI capabilities of most countries, the majority of discussions in the studied international forums are also in their early stages. Like other emerging technologies and their regulations, countries are racing to catch up with the ever-evolving technologies and technological complexities and AI is no exception; policymakers are still debating over the potential impacts and challenges related to AI [56,57]. It is observed that while the focus of most national strategies is promoting domestic innovation and strengthening national capacities, the UNESCO recommendations place more emphasis on SDGs, human rights, privacy, and accountability (Sections 3 and 5). This disconnect is understandable, given that the priorities and mandates of different organisations may be dissimilar; however, in order to incorporate a global framework that is truly germane and adoptable for all countries, policymakers must direct more efforts towards reducing these disconnects and discrepancies by harmonisation of national and global priorities.
Notwithstanding these discrepancies, there are two important policy recommendations that pre-empt the regularisation of global strategies: i) Although there exist international platforms like GPAI, OECD AI Policy Observatory, and other such platforms that engage and expand the discourse surrounding AI regulation and governance, only select countries are part of these platforms. To ensure uniformity and congruence within local and global strategies, a truly global and inclusive platform for AI regulation and governance is the need of the hour. The regulation and governance is especially required for ensuring that appropriate safety, security and privacy measures are incorporated in the new AIbased digital platforms that are being developed. ii) The drafting of such a strategy and a set of rules and standards must be grounded in the inputs from multiple stakeholders, including local governments, academia and private industry across all countries irrespective of their capabilities and ranking. In this regard, platforms like GPAI, as a specialised international forum on AI, can provide crucial research and policy support by being more inclusive. However, given the possibilities with AI and the effects it will have on all the countries across the globe, a truly global platform would be needed, and a special UN framework convention on AI might be the right way forward. Here, diplomats and policymakers should take leadership in negotiating such a multilateral convention based on agreeable values and principles.
Conclusion
Amid the rising importance and investment in AI, it has an inevitable influence on both diplomatic discussions and the practice of diplomacy. AI was presented and analysed as an area of international cooperation based on the initiatives of key countries and international forums. It was noted that AI international policy discourse and national strategies for international cooperation-are still in early phases. In the future, AI governance and regulation is expected to be affected by the larger geopolitical discourse. This is because AI is viewed as a critical, strategic technology due to the platforms that can be built using it as well as its economic and military implications. Multilateral forums like GPAI, UNESCO and G20 can play key roles in defining the international rules, norms and standards related to AI and large multinational technology companies are likely to play critical roles defining the contours of AI development; engaging the technology sector, such as through regulatory and incentivized measures, to achieve the national and international AI policy objectives will continue to be a challenge. Overall, AI is already viewed as a crucial matter for diplomatic discussions but whether, in the future, these will be done in the spirit of either cooperation or competition remains to be seen.
References
[1] Roberts, Megan. 2021. “International Cooperation on Digital and Tech: Outlook for The Rest Of 2021.” United Nations Foundation. 2021. https://unfoundation.org/blog/post/international-cooperation-on-digitaland-tech-outlook-for-the-rest-of2021/
[2] Scott, Ben, Stefan Heumann, and Philippe Lorenz. 2018. “Artificial Intelligence and Foreign Policy,” no. January.
[3] Diplo. 2021. “Ai and Diplomacy.” 2021. https://www.diplomacy.edu
[4] Giri, Chaitanya, and Amrita Menon. 2020. “NEST-Ling New Tech in Foreign Policy.” 2020.
https://www.gatewayhouse.in/nest-tech-foreign-policy/
[5] Ray, Trisha, and Akhil Deo. 2020. “Priorities for a Technology Foreign Policy for India,” no. 403: 1–20.
[6] Horowitz, Michael. 2011. “Artificial Intelligence, International Competition, and the Balance of Power.” Quarterly Journal of Economics 126 (3): 1133–72.
[7] F. Kerry, Cameron, Joshua P. Meltzer, Andrea Renda, Alex Engler, and Rosanna Fanni. 2021. “Strengthening International Cooperation on AI.” https://www.brookings.edu/articles/strengthening-international-cooperation-on-ai/
[8] Tiberghien, Yves, Danielle Luo, and Panthea Pourmalek. 2021. “Existential Gap: Digital/AI Acceleration and the Missing Global Governance Capacity.” Project for Peaceful Competition. 2021. https://www.peaceful-competition.org/pub/85cu1r9n/release/1
[9] Kapetas, Anastasia. 2020. “The Geopolitics of Artificial Intelligence.” 2020.
[10] Børge Brende. 2022. “These Are the Stepping Stones towards Global Cooperation.” World Economic Forum. https://www.weforum.org/agenda/2022/01/these-are-the-stepping-stones-needed-towards-global-cooperation/
[11] Wagner, Caroline S., Han Woo Park, and Loet Leydesdorff. 2015. “The Continuing Growth of Global Cooperation Networks in Research: A Conundrum for National Governments.” Edited by Wolfgang Glanzel. PLOS ONE 10 (7): e0131816. https://doi.org/10.1371/journal.pone.0131816
[12] Erdélyi, Olivia J., and Judy Goldsmith. 2018. “Regulating Artificial Intelligence.” In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, 95–101. New York, NY, USA: ACM. https://dl.acm.org/doi/10.1145/3278721.3278731
[13] Luengo-Oroz, Miguel, Katherine Hoffmann Pham, Joseph Bullock, Robert Kirkpatrick, Alexandra Luccioni, Sasha Rubel, Cedric Wachholz, et al. 2020. “Artificial Intelligence Cooperation to Support the Global Response to COVID-19.” Nature Machine Intelligence 2 (6): 295–97. https://doi.org/10.1038/s42256-020-0184-3
[14] The World Bank. 2023. “The World by Income and Region.” World Bank.
https://datatopics.worldbank.org/world-development-indicators/the-world-by-income-and-region.html
[15] CIA. 2023. “The World Factbook.” https://www.cia.gov/the-world-factbook/
[16] United Nations Conference on Trade and Development. 2021. Technology and Innovation Report 2021. https://unctad.org/page/technology-and-innovation-report-2021
[17] Intueri Consulting, 2021. A comparative study of China and India’s AI policy.
https://intueriglobal.com/a-comparative-study-of-china-and-indias-ai-policy/
[18] Tortoise. 2021. “The Global AI Index.” https://www.tortoisemedia.com/intelligence/global-ai/
[19] OECD. 2022. “National AI Policies & Strategies.” https://oecd.ai/en/dashboards
[20] WIPO. 2019. “WIPO Technology Trends 2019 – Artificial Intelligence.” https://www.wipo.int/publications/en/details.jsp?id=4386
[21] Johnson, R. B., & Onwuegbuzie, A. J. 2004. Mixed methods research: A research paradigm whose time has come. Educational researcher, 33(7), 14-26.
[22] Datawrapper. 2022. “Datawrapper.” 2022. https://www.datawrapper.de
[23] Worldometer. 2022. “Worldometer.” 2022. https://www.worldometers.info
[24] WIPO. 2019. The Global Innovation Index 2019. World Intellectual Property Organization. https://www.wipo.int/edocs/pubdocs/en/wipo_pub_gii_2019-chapter1.pdf
[25] Niti Aayog. 2018. “National Strategy for Artificial Intelligence.” New Delhi. https://indiaai.gov.in/documents/pdf/NationalStrategy-for-AI-Discussion-Paper.pdf
[26] Roy, Anna. 2021. “Responsible Ai #AIForAll Part 1.” NITI Aayog, no. February 8. https://indiaai.gov.in/research-reports/responsible-ai-part-1-principles-for-responsible-ai
[27] NITI Aayog. 2021. “Responsible AI #AIForAll Part 2.” NITI Aayog, no. August 17. https://indiaai.gov.in/research-reports/responsible-ai-part-2-operationalizing-principles-for-responsible-ai/
[28] Sibal, Kanwal. 2022. “Is India Muscling Its Strategic Autonomy by Engaging Both the East and the West?” India Narrative. https://www.indianarrative.com/opinion-news/is-india-muscling-its-strategic-autonomy-by-engaging-both-the-east-and-the-west-38333.html
[29] Webster, Graham, Rogier Creemers, Paul Triolo, and Elsa Kania. 2017. “China’s ’New Generation Artificial Intelligence Development Plan.” New America. https://www.newamerica.org/cybersecurity-initiative/digichina/blog/full-translation-chinas-new-generation-artificial-intelligence-development-plan-2017/
[30] Webster, Graham, Rogier Creemers, Paul Triolo, and Elsa Kania. 2017. “China’s Plan to ‘Lead’ in AI: Purpose, Prospects, and Problems.” New America. https://www.newamerica.org/cybersecurity-initiative/blog/chinas-plan-lead-ai-purpose-prospects-and-problems/
[31] Shenzhen Government. 2021. “Regulations of Shenzhen Special Economic Zone on National Independent Innovation Demonstration Zone.” Shenzhen Government Online.
http://www.sz.gov.cn/en_szgov/laws/content/post_8860145.html
[32] Iyengar, Rishi. March 2023. “Who’s Winning the AI Race? It’s Not That Simple.” Foreign Policy, https://foreignpolicy.com/2023/03/27/us-china-ai-competition-cooperation/
[33] Westphal, Theo, and Ruoxi Wang. 2023. “Artificial Intelligence with Chinese Characteristics: National Strategy, Security and Authoritarian Governance.” International Affairs 99(1): 410–11. https://doi.org/10.1093/ia/iiac269
[34] Mari, Angelica. 2020. “Brazil Creates National AI Innovation Network.” ZD Net. https://www.zdnet.com/article/brazil-creates-national-ai-innovation-network/
[35] McFarland, Alex. 2021. “Brazil Establishes National AI Strategy.” Unite.AI.
https://www.unite.ai/brazil-establishes-national-ai-strategy/
[36] Evangelos, Sakiotis, Meneses Anna Oberschelp de, Shepherd Nicholas, and Quathem Kristof Van. 2023. “Brazil’s Senate Committee Publishes AI Report and Draft AI Law.” Covington. 2023.
[37] The Federal Government. 2020. “Artificial Intelligence Strategy of the German Federal Government.”
https://www.ki-strategie-deutschland.de/files/downloads/Fortschreibung_KI-Strategie_engl.pdf
[38] Executive Office of the President. 2019. “Maintaining American Leadership in Artificial Intelligence.” Vol. 84. https://www.govinfo.gov/content/pkg/FR-2019-02-14/pdf/2019-02544.pdf
[39] The White House Office of Science and Technology Policy. 2020. “American Artificial Intelligence Initiative: Year One Annual Report,” no. February: 36.
https://trumpwhitehouse.archives.gov/wp-content/uploads/2020/02/American-AI-Initiative-One-Year-Annual-Report.pdf
[40] GPAI. 2022. “About GPAI.” https://gpai.ai/about/
[41] OECD. 2022. “OECD AI Principles Overview.” https://oecd.ai/en/ai-principles
[42] OECD. 2022. “International Co-Operation for Trustworthy AI.” https://oecd.ai/en/dashboards/ai-principles/P14
[43] UNESCO. 2022. “Recommendation on the Ethics of Artificial Intelligence.” https://unesdoc.unesco.org/ark:/48223/pf0000381137
[44] UNESCO. 2021. “UNESCO Member States Adopt the First Ever Global Agreement on the Ethics of Artificial Intelligence.” https://www.unesco.org/en/articles/unesco-member-states-adopt-first-ever-global-agreement-ethics-artificial-intelligence
[45] G7. 2017. “G7 ICT and Industry Ministers’ Declaration: Making the Next Production Revolution Inclusive, Open and Secure,” no. September: 25–26. http://www.g7.utoronto.ca/ict/2017-ict-declaration.html
[46] G7. 2018. “Chairs’ Summary: G7 Ministerial Meeting on Preparing Jobs of the Future.”
https://www.international.gc.ca/world-monde/international_relations-relations_internationales/g7/documents/2018-03-27-chairs_summary-resume_presidents.aspx?lang=eng
[47] G7. 2019. “Biarritz Strategy for an Open, Free and Secure Digital Transformation,” no. August: 4–5. https://g7g20-documents.org/database/document/2019-g7-france-leaders-leaders-language-biarritz-strategy-for-an-open-free-and-secure-digital-transformation
[48] US Department of State. 2020. “G7 Science and Technology Ministers’ Declaration on COVID-19.” https://www.state.gov/g7-science-and-technology-ministers-declaration-on-covid-19/
[49] G20. 2019. “G20 Osaka Leaders’ Declaration.” https://www.mofa.go.jp/policy/economy/g20_summit/osaka19/en/documents/final_g20_osaka_leaders_declaration.html
[50] G20. 2019. “G20 AI Principles,” 1–4. https://www.g20-insights.org/related_literature/g20-japan-ai-principles/
[51] AI Policy. 2021. “The Quad Critical and Emerging Technologies Working Group: Tech for an Open and Inclusive Indo-Pacific.” https://aipolicy.xyz/international-matters/2021/03/the-quad-critical-and-emerging-technologies-working-group-tech-for-an-open-and-inclusive-indo-pacific/
[52] Ministry of External Affairs. 2021. “Fact Sheet: Quad Leaders’ Summit.”
https://mea.gov.in/outoging-visit-detail.htm?34319/Fact+Sheet+Quad+Leaders+Summit
[53] Australia National University. 2022. “Quad Tech Network.”
https://nsc.crawford.anu.edu.au/department-news/18328/quad-tech-network
[54] Kwet, Michael. 2019. “Digital Colonialism: US Empire and the New Imperialism in the Global South.” Race & Class 60 (4): 3–26. https://doi.org/10.1177/0306396818823172
[55] Sey, Araba, Gabriella Razzano, Andrew Rens, Naila Govan Vassen, Shamira Ahmed, Oarabile Mudongo, and Alison Gillwald. 2021. “Mapping Policy and Capacity for Artificial Intelligence for Development in Africa.” https://healthtechafrica.org/wp-content/uploads/2022/02/Mapping-AI-capacity.pdf
[56] OECD. 2021. The Digital Transformation of SMEs. OECD Studies on SMEs and Entrepreneurship. OECD. https://www.oecd.org/industry/smes/PH-SME-Digitalisation-final.pdf
[57] Misra, Santosh K., Satyasiba Das, Sumeet Gupta, and Sujeet K. Sharma. 2020. “Public Policy and Regulatory Challenges of Artificial Intelligence (AI).” In , 100–111. https://doi.org/10.1007/978-3-030-64849-7_10
About the Authors
Amruta Mahuli
Amruta is an incoming doctoral candidate at the Max Planck Institute for Security and Privacy. She holds a Bachelors degree in Computer Science (BSc Computer Science) and law (LLB) and has completed her Masters (MA) in Public Policy and Governance. She has work experience as a data analyst, legal professional specializing in intellectual property, contract drafting and negotiation, and research intern for several technology policy based think tanks. Her research interests are grounded in the intersection of law, emerging technologies, and the welfare of the marginalized.
Corresponding address: amruta.u.mahuli@gmail.com
Sumit Kumar
Sumit Kumar is a Doctoral Researcher at Monash University. His research project is focused on understanding the current technological innovation in the field of Artificial Intelligence targeted toward accelerating Net Zero Energy Transition. He has pursued a masters in Sustainable Development from Science Policy Research Unit (SPRU), University of Sussex with a specialisation in science, technology and innovation policy. He is the recipient of Chevening Scholarship (Funded by the Foreign, Commonwealth and Development Office, UK Government) and Sussex India Scholarship. In his previous professional engagement he was associated with the Indo US Science and Technology Forum (IUSSTF) as part of the US India Artificial Intelligence Initiative (USIAI). Prior to pursuing higher education in the UK, he was associated with J-PAL South Asia (MIT, Boston) and the Ministry of Rural Development, Government of India.
ORCID: 0009-0001-6952-1723
Corresponding address: sumit.kumar@monash.edu
Suryesh K Namdeo
Suryesh K Namdeo is a visiting scholar at the DST Centre for Policy Research at Indian Institute of Science, Bengaluru. He works in the areas of science
diplomacy and biosecurity. He is a member of the Indian National Young Academy of Sciences (INYAS) and the director of outreach and engagement for the Journal of Science Policy & Governance (JSPG). He has worked as a science & technology consultant with the United Nations and holds a PhD from the Max Planck Institute for Biology, Tuebingen, Germany.
Corresponding address: suryeshn@iisc.ac.in
Conflict of interest: The authors declare no conflict of interest.
Glossary
AI – Artificial Intelligence
CAGR – Compound Annual Growth Rate
CSTS – Centre for Studies on Technological Sustainability
EU – European Union
G7 – The Group of Seven
G20 – The Group of Twenty
GDPR – General Data Protection Regulation
GDP – Gross Domestic Product
GII – Global Innovation Index
Download the Appendices