Provider of in-depth & balanced analysis on hot issues around the world

Beijing’s Victory Day parade in Tiananmen Square was designed to dazzle: ranks of uniformed troops, formations of aircraft, and an arsenal of new systems meant to underscore China’s rapid military modernization. But the most consequential image was not a missile or a stealth jet. It was a tableau of three leaders—Xi Jinping at the center, flanked by Vladimir Putin and Kim Jong Un—watching the spectacle together. The scene, widely broadcast and photographed, turned a commemorative event into a geopolitical marker. It was less a snapshot than a signal: the public normalization of a deepening alignment among China, Russia, and North Korea, at a moment when Western democracies are struggling to sustain cohesion on core strategic questions. The parade itself offered the familiar mixture of hardware and narrative. Coverage highlighted the unveiling or public confirmation of advanced systems across domains: upgraded intercontinental missiles, new submarine-launched ballistic missiles, hypersonic and anti-ship capabilities, long-range bombers, early warning aircraft, and a broad stable of unmanned platforms, including undersea vehicles and “loyal wingman” drones. Chinese media presented these developments as evidence of a “world-class” People’s Liberation Army (PLA) moving beyond legacy constraints and into truly multi-domain operations, with information, space, and cyber now integrated alongside land, sea, and air. Independent reporting catalogued the breadth of systems and emphasized a narrative of credible deterrence and strategic depth rather than mere choreography. Yet the more instructive message was political. The presence of Putin and Kim, alongside other leaders, was not a mere ceremonial occurrence. Each leader arrived with clear incentives to be seen at Xi’s side, and each gained by lending visual weight to Beijing’s story. For Moscow, the image reinforced the claim that Russia is not isolated, that it retains powerful partners and is embedded in a wider non-Western coalition. For Pyongyang, the moment was even more significant: an opportunity to step out of diplomatic isolation and be recognized publicly as a member of a consequential strategic grouping. For Beijing, hosting both leaders signaled that China can convene and coordinate—projecting status, reassuring sympathetic governments, and unsettling adversaries by hinting at a tighter web of cooperation among U.S. rivals. The convergence behind the optics has been building for years, and could have happened only on Chinese soil. China and Russia have expanded their coordination across energy, defense, and diplomatic, even as they preserve maneuvering room on sensitive issues. North Korea’s accelerating exchanges with Russia, alongside growing political warmth with Beijing, provide a third leg to this emerging tripod. None of this amounts to a formal alliance with mutual defense obligations. But it does resemble a strategic alignment held together by shared interests: resisting a U.S.-led order, blunting sanctions pressure, reducing vulnerability to Western technology restrictions, and demonstrating that alternatives exist to dollar-centric finance and Western supply chains. The choreography on the rostrum did not create this alignment; it made it more legible and clear. Memory politics is a key component of that legibility. Beijing’s decision to anchor the parade in the commemoration of victory over Japan allows contemporary power projection to be cloaked in a unifying moral narrative. China increasingly leverages World War II memory in diplomacy—shaping a “memory war” that reframes the post-1945 order and what is seen from China as its rightful place within it. Russia’s long-standing use of the “Great Patriotic War” plays a parallel role, justifying current policies through selective historical continuity. North Korea’s revolutionary mythology fits easily into this narrative architecture. By standing together at an anniversary of anti-fascist victory, the three leaders signaled an ideational convergence that complements their material cooperation: a claim to moral legitimacy as guardians of an alternative international vision. The military dimension of the parade, while not the core of this argument, still matters. Displays of a maturing triad—land-based ICBMs, submarine-launched systems, and an air-launched nuclear component—aim to convey survivable second-strike capacity. The public presentation of hypersonic and anti-ship systems is meant to complicate adversary planning in the Western Pacific. The range of unmanned platforms suggests an intent to saturate domains with relatively low-cost, attritable assets, improving persistence and compressing the sensor-to-shooter loop. It is prudent to treat parades cautiously: not all showcased systems are fully operational or fielded at scale, and performance claims are difficult to validate. But as an indicator, the breadth and integration of platforms reflect a planning culture committed to joint operations and “intelligentized” warfare, where AI-enabled targeting and decision support are not theoretical ambitions but programmatic priorities What, then, does the image of Xi–Putin–Kim actually change? First, it clarifies expectations. Observers no longer need to infer the trajectory of this triangular relationship from scattered bilateral overtures. The three leaders have chosen to make their alignment visible. Visibility creates deterrent value, raising the perceived costs of coercing any one member, and it can also facilitate practical cooperation: intelligence sharing, diplomatic coordination at the UN and other fora, synchronized signaling during regional crises, and mutually reinforcing sanctions-evasion practices. Second, it complicates Western planning. Even if Beijing keeps caution around direct military assistance in Europe or the Korean Peninsula, diplomatic top-cover, economic buffering, and technology flows short of lethal aid can still alter the correlation of forces over time. Finally, it resonates across the Global South. Many governments seek strategic autonomy and resist being forced into binary choices. The parade’s optics supplied a ready-made narrative for those who argue that the international system is already multipolar and that non-Western coalitions can deliver security and development without Western tutelage. The contrast with Western coordination was strikingly evident. In the transatlantic community, support for Ukraine remains substantial; however, debates about resource levels, war aims, and timelines have intensified. In the Indo-Pacific, there is a growing alignment on deterring coercion in the Taiwan Strait and the South China Sea; however, national economic interests and differing risk tolerances result in uneven policies toward China. Across Europe and North America, electoral politics continue to inject volatility into foreign policy, complicating efforts to sustain long-term, bipartisan strategies. None of these frictions amounts to collapse, and there are genuine Western successes in coalition-building—from NATO enlargement to evolving minilateral formats in the Indo-Pacific. However, an analytically honest reading of the moment acknowledges that the authoritarian trio in Beijing has projected a unity of purpose that Western capitals currently struggle to match consistently. Three implications follow. The first is narrative competition. If Beijing, Moscow, and Pyongyang can turn a commemorative event into a global story about legitimacy and resilience, they will continue to use history as a strategic resource. The appropriate Western response is not to cede the narrative field but to invest in historically grounded, forward-looking messaging that explains the link between rules-based order and practical benefits—trade reliability, crisis management, and sovereignty protection—for diverse audiences. The second is coalition maintenance. Western policymakers will need to prioritize “coalition hygiene”: aligning export controls and investment screening where it matters most; building redundancy into critical supply chains; closing divergences in sanctions enforcement; and coordinating messaging so that tactical differences do not obscure strategic alignment. This requires political discipline more than new institutions. The third is theater integration. As the Beijing image suggested a cross-regional understanding among three adversarial capitals, allied planning must better account for cross-theater linkages—how actions in Europe affect deterrence in Asia, and vice versa—and ensure that resource allocations and industrial policies reflect genuinely global prioritization. It is important not to overstate. The emerging alignment among China, Russia, and North Korea is asymmetric and interest-based, not a tightly binding alliance. Beijing’s global economic integration imposes constraints that Moscow and Pyongyang do not share. Russia and North Korea each bring liabilities that China will manage carefully. Frictions—over technology, pricing, and regional equities—will persist. But the threshold crossed in Beijing is nonetheless meaningful. These governments judged that the benefits of public proximity now outweigh the costs. That judgment, once made, is difficult to reverse quickly; it tends to generate its own momentum through bureaucratic follow-through and sunk reputational costs. One image cannot rewrite the balance of power. It can, however, crystallize a trend and concentrate minds. The sight of Xi, Putin, and Kim standing together did exactly that. It captured an authoritarian convergence rooted in shared grievances and converging strategies, and it highlighted the challenge facing democracies that wish to preserve an open and stable order: maintaining the patience, unity, and policy discipline to act together. The test for the West is less whether it recognizes the signal—most capitals do—than whether it can convert recognition into sustained, collective action. If Beijing’s parade was a demonstration of choreography and intent, the appropriate answer is not a counter-parade, but the quieter work of alignment: aligning narratives with interests, interests with instruments, and instruments with partners. That work is not glamorous. It is, however, what turns a photo into policy.

Introduction Geopolitical tensions are making global trade increasingly difficult. In order to reduce the associated risk of default, companies are shifting their trade relations to trading partners that are politically similar to them. In the course of the beginnings of geo-economic fragmentation, politically and economically like-minded countries are also gaining in importance for German and European decision-makers. Liaison countries1 in particular can form a counterforce to the trend towards polarization in foreign trade - especially between the USA and China: they are characterized by a pronounced economic and trade policy openness that overrides differences between geopolitical or ideological camps. Consequently, the question arises: How can relevant connecting countries for Germany and Europe be identified? What opportunities and risks do closer trade relations with these countries offer in order to strengthen foreign trade resilience in geopolitically uncertain times?  With a high degree of openness - defined as the sum of imports and exports in relation to gross domestic product - of over 80 percent2 , the German economy is strongly integrated into global trade. Accordingly, the disruptive effect of geo-economic fragmentation on the German economy would be above average. The defensive strategy to strengthen Germany's economic security by pushing for trade policy independence would only reinforce geo-economic fragmentation. Against the backdrop of comparatively high economic vulnerability, it is necessary to focus on those potential partner countries with which German and European foreign trade could be developed and expanded even under the condition of increasing fragmentation.  Geoeconomic Fragmentation  The term "geo-economic fragmentation" is used to describe the politically motivated reorganization of global goods and financial flows, in which strategic, economic and political interests primarily determine the choice of countries of origin and destination for trade flows.3 In the scenario of geo-economic fragmentation, the result would be the formation of a bloc within the global community of states, which would fundamentally change the regulatory structure of global economic networking. In this case, trade and investment would probably concentrate from a previously diverse range of economic partner countries - prior to the formation of the bloc - on those countries that now - since the formation of the bloc - belong to the same bloc.  The likelihood of this scenario occurring and leading to an increased fragmentation of the global economic order has increased again in the recent past. For example, Donald Trump's second term as US president is causing increasing geopolitical uncertainty worldwide.  Statements on the concrete form of a possible demarcation of potential blocs are subject to a great deal of uncertainty. However, the division of a large part of the global economy into a "US bloc" and a "China bloc" is a conceivable scenario for which German politics and business should prepare.  Data already shows that, at a global level, foreign trade openness has decreased in the recent past. Data from the World Trade Organization (WTO) illustrates the increasing hurdles in global trade in goods. While 3.1% of global imports were still affected by tariff or non-tariff barriers to trade in 2016 - including under WTO rules - this figure rose to 11.8% in 2024 over the following years.4 This development goes hand in hand with a noticeable loss of importance and enforcement of the WTO since the 2010s, which previously played a central role as the guardian of the rules-based global economic order.  Studies by the International Monetary Fund (IMF) have already found indications of an incipient geo-economic fragmentation along potential bloc borders. It shows that trade in goods and foreign direct investment between countries that would belong to the opposing camp in the event of a bloc formation declined on average in 2022 and 2023 - in contrast to foreign trade between countries that are geopolitically close.5  In this initial phase of geo-economic fragmentation, liaison countries are beginning to establish themselves as a counterforce, holding the fragmenting global community of states together with new trade and investment routes.  Identification of liaison countries Specifically, liaison countries have the following characteristics: a pronounced openness to foreign trade in the form of a high foreign trade quota and low tariff and non-tariff trade barriers, as well as pronounced economic relations with partner countries from different geopolitical camps. The geopolitical orientation of countries can be examined using data on voting behavior within the United Nations.6 This involves analyzing whether a country can be assigned to the US or Chinese camp - or whether there is no pronounced proximity and therefore political neutrality or "non-alignment" in the sense of ideological independence. The data-based identification of connecting countries is relatively new. Empirical analyses are also limited to connecting countries in the context of US-Chinese foreign trade - specifically US imports from China. In this case, the characteristics of a connecting country can be broken down into (1) "non-alignment" - i.e. a geopolitical distance to both a Western and an Eastern bloc - as well as (2) an increase in imports and foreign investment from China and (3) a simultaneous increase in exports to the United States. In a narrower sense, this is an evasive reaction to trade restrictions, i.e. circumventing trade. If the foreign trade indicators - specifically the trade and investment data relating to the US and China - of "non-aligned" countries for the period from 2017 to 2020 show corresponding characteristic-related changes compared to previous years, these can be identified as countries connecting the US and China.  The analysis of trade data shows that the value of direct exports from China to the USA fell during Donald Trump's first term in office. At the same time, both Chinese exports to some of the "non-aligned" countries and exports from these countries to the USA have increased significantly. These countries have presumably stepped in as a link on the export route from China to the US after the previously direct trade flow was interrupted by trade barriers and had to find a new route. Companies producing in China are therefore likely to have sought new, indirect ways to maintain access to the US sales market.  A certain statistical inaccuracy in the foreign trade data makes it difficult to draw a definitive conclusion in this context. It should be noted: No single commodity can be tracked across national borders in trade data collection. Whether the additional goods imported from China actually found their way to the United States can only be assumed approximately. However, if the trade flows are aggregated, a clearer picture emerges and the circumvention trade via selected connecting countries - including Vietnam and Mexico - becomes visible.  Data on foreign direct investment rounds off the analysis.7 "Non-aligned" countries in which an increase in Chinese investment can be seen between 2016 and 2020 in addition to trade flows can be identified as connecting countries. Here, too, available data suggests that the companies concerned either exported their goods to the United States via a stopover or even outsourced parts of their production destined for the US market to connecting countries. Five connecting countries between the US and China Based on the 2017-2020 study period, various connecting countries can be empirically identified that were used to indirectly maintain access to the US market. In terms of foreign trade volume, the economically most important connecting countries include Mexico, Vietnam, Poland, Morocco and Indonesia.8 All five countries are characterized by the fact that both their exports of goods to the US and their imports of goods from China increased significantly between 2017 and 2020. In addition, greenfield investments (foreign direct investment to set up a new production facility) have risen significantly compared to the period before 2017.  However, the five countries show different priorities in their development, which differentiate them in their role as connecting countries between the USA and China. In Vietnam, exports to the USA in particular have risen sharply. China has been the most important procurement market for Vietnamese companies for years. Poland, Mexico and Indonesia are characterized as connecting countries primarily by the significant increase in imports from China. Morocco, in turn, was able to attract more Chinese foreign investment in particular. Greenfield investments have almost tripled here since 2017. However, Poland - a rather surprising candidate for the role of liaison country, as it is intuitively assigned to the US-oriented bloc - is positioned fairly centrally between the US and China according to the analysis of voting behavior within the United Nations9. In addition, Poland qualifies primarily due to the sharp rise in greenfield investments from China, primarily in the expansion of domestic battery production.10  It cannot be concluded from the previous studies on the USA and China whether German companies are also circumventing trade barriers from the USA via the countries identified. As the trade policy conflicts between the US and China differ significantly from those between the EU and China, there has been a lack of comparable empirical data to analyze connecting countries in the EU context. Opportunities and challenges As the German economy is strongly oriented towards foreign trade and is closely networked with both the USA and China, German companies play a particularly exposed role in the area of tension between the USA and China. Increased economic exchange with potential connecting countries would offer German companies an opportunity to mitigate the expected shock of a geopolitical bloc. They could at least maintain international trade to a certain extent and thus secure some of the endangered sales and procurement markets. On the other hand, there are also costs associated with expanding foreign trade relations with potential connecting countries. The greater complexity also increases the risk in the value chains. Companies that position themselves wisely within this trade-off buy themselves valuable time in the event of a shock to reorganize themselves against the backdrop of changed foreign trade conditions.  From the perspective of foreign trade policy, it is also possible to examine the extent to which stronger foreign trade cooperation with (potential) connecting countries could have advantages. The trade-off between resilience and complexity must then be assessed at a macroeconomic level, beyond individual company interests. In order to make it easier for companies to connect to potential connecting countries and to create appropriate framework conditions, German and European policy can build on existing comprehensive strategies at national and European level. Both the China Strategy11 and the National Security Strategy12 focus foreign policy on connecting countries as part of a stronger economic and political risk diversification. There is also a similar framework at European level with the EU's Strategic Compass13 . Following on from this, the German government could create targeted incentives to open up new markets in liaison countries, which would diversify critical supply chains and reduce one-sided dependencies.  At the same time, connecting countries pose a challenge. These can be used to circumvent foreign trade measures such as sanctions if flows of goods can find alternative routes via connecting countries more easily than before.  In order to realize opportunities and overcome challenges, close cooperation between science, politics and companies is required. This first requires the identification of a selection of potential connecting countries through scientifically sound analysis. This creates the basis for the subsequent steps in which European and German policymakers work closely with companies to create attractive framework conditions for trade with potential connecting countries - for example through bilateral trade agreements.  Attractive foreign trade framework conditions can create the necessary incentive to actually expand trade relations with potential connecting countries. Companies need to weigh up individual cases and make forward-looking decisions: To what extent is there a risk of a loss of production triggered by geopolitical conflicts? And how much would the complexity of the value chain increase if more potential connecting countries were included? Ultimately, the actual choice of preferred sales and procurement markets lies with the individual companies. LicenseThis work is licensed under CC BY 4.0 References1. Verbindungsländer werden im Sinne von Connectors verstanden, vgl. Gita Gopinath/Pierre-Olivier Gourinchas/Andrea F Presbitero/Petia Topalova, Changing Global Linkages: A New Cold War?, Washington, D.C.: IMF, April 2024 (IMF Working Paper) <https://www.imf.org/en/Publications/WP/Issues/2024/04/05/Changing-Global-Linkages-A-New-ColdWar-547357/>. 2. Statistisches Bundesamt (Destatis), Außenwirtschaft. 2025, <https://www.destatis.de/DE/Themen/Wirtschaft/Globalisierungsindikatoren/aussenwirtschaft.html#246 078/>.  3. Shekahar Aiyar/Franziska Ohnsorge, Geoeconomic Fragmentation and ‚Connector’ Countries, Online verfügbar unter:  <https://mpra.ub.uni-muenchen.de/121726/1/MPRA_paper_121726.pdf>.4. WTO, WTO Trade Monitoring Report, Genf, November 2024, <https://www.wto.org/english/tratop_e/tpr_e/factsheet_dec24_e.pdf/>. 5. Gita Gopinath/Pierre-Olivier Gourinchas/Andrea F Presbitero/Petia Topalova, Changing Global Linkages: A New Cold War?, Washington, D.C.: IMF, April 2024 (IMF Working Paper) <https://www.imf.org/en/Publications/WP/Issues/2024/04/05/Changing-Global-Linkages-A-New-ColdWar-547357/>.  6. Michael A. Bailey/Anton Strezhnev/Erik Voeten, »Estimating Dynamic State Preferences from United Nations Voting Data«, Journal of Conflict Resolution, 61 (2017) 2, S. 430-456, <https://journals.sagepub.com/doi/10.1177/0022002715595700/>.7. Gita Gopinath/Pierre-Olivier Gourinchas/Andrea F Presbitero/Petia Topalova, Changing Global Linkages: A New Cold War?, Washington, D.C.: IMF, April 2024 (IMF Working Paper) <https://www.imf.org/en/Publications/WP/Issues/2024/04/05/Changing-Global-Linkages-A-New-ColdWar-547357/>. War-547357. 8. Enda Curran/Shawn Donnan/Maeva Cousin, »These Five Countries are Key Economic ‚Connectors‘ in a Fragmenting World«, in Bloomberg (online), 1.11.2023, <https://www.bloomberg.com/news/articles/2023-1102/vietnam-poland-mexico-morocco-benefit-from-us-china-tensions/>.9. Michael A. Bailey/Anton Strezhnev/Erik Voeten, »Estimating Dynamic State Preferences from United Nations Voting Data«, Journal of Conflict Resolution, 61 (2017) 2, S. 430-456, <https://journals.sagepub.com/doi/10.1177/0022002715595700/>.  10. Enda Curran/Shawn Donnan/Maeva Cousin, »These Five Countries are Key Economic ‚Connectors‘ in a Fragmenting World«, in Bloomberg (online), 1.11.2023, <https://www.bloomberg.com/news/articles/202311-02/vietnam-poland-mexico-morocco-benefit-from-us-china-tensions/>.11. Auswärtiges Amt, China‐Strategie der Bundesregierung, Berlin, Juli 2023, <https://www.auswaertigesamt.de/resource/blob/2608578/810fdade376b1467f20bdb697b2acd58/china-strategie-data.pdf/>.  12. Auswärtiges Amt, Integrierte Sicherheit für Deutschland: Nationale Sicherheitsstrategie, Berlin, Juni 2023, <https://www.bmvg.de/resource/blob/5636374/38287252c5442b786ac5d0036ebb237b/nationalesicherheitsstrategie-data.pdf/>.  13. Rat der Europäischen Union, Ein Strategischer Kompass für Sicherheit und Verteidigung, Brüssel, März 2022, <https://data.consilium.europa.eu/doc/document/ST-7371-2022-INIT/de/pdf/>.

As AI accelerates, the US and China shape the global governance landscape through parallel ambitions and contrasting ideologies. Understanding their policy trajectories reveals key inflection points for potential engagement. The United States and China are the world’s leading powers in artificial intelligence (AI)—and each has global ambitions. As AI development accelerates, so too do calls to regulate it responsibly. Both countries have the capacity to shape the future of AI governance. But understanding where cooperation might be possible requires understanding how each country’s approach has evolved. While their methods and ideologies differ, and their leadership aspirations often appear at odds, two underexamined forces—infrastructure and philosophy—may create unexpected space for mutual engagement. In the US, AI governance has long been shaped by the mythology of the free market. Under President Barack Obama, that mythology was tempered by calls for pipeline diversity. President Donald Trump discarded these in favour of rhetoric about “American values,” an ill-defined phrase deployed more as a competitive cudgel against China than a coherent policy vision. President Joe Biden attempted to resurrect some of Obama’s normative commitments—launching the Blueprint for an AI Bill of Rights, Executive Order 14110 on “safe, secure, and trustworthy AI,” and initiatives like the National AI Research Resource—but the foundations of his administration’s policies remained countering China’s influence through increased export controls and other measures. Trump’s second administration has once again changed focus away from equity and community-centredness, but the continued focus on competing with China and ensuring US “victory” in the “AI race” shows there’s more continuity than often assumed between different administrations’ AI approaches. China, for its part, has been more consistent in its governance strategy, even as it experiments at the margins. Its 2017 New Generation AI Development Plan cast AI as a pillar of national strength, and subsequent regulations on algorithms, deepfakes, generative AI, and facial recognition underscore a clear priority: balancing the “twin miracles” of economic development and social stability. Though early days of “fragmented authoritarianism” have given way to a more centralised approach with specific vertical laws, this logic remains embedded throughout the governance system. China is generally less explicit in its rhetoric than the US, with fewer exhortations about race dynamics and countering the US. In fact, it frequently invokes ideas of ethical pluralism and the language of international cooperation. This framing likely reflects a blend of genuine ideological positioning and diplomatic strategy, but it presents a challenge to US policymakers who continue to frame AI governance in moralising binaries. This is evident in two parallel AI Action Plans released in July 2025. The US released a guiding document—indicatively titled “Winning the Race: America’s AI Action Plan”—with an epigraph from President Trump stating: “As our global competitors race to exploit these technologies, it is a national security imperative for the United States to achieve and maintain unquestioned and unchallenged global technological dominance.” Though the Action Plan itself contains only three substantive references to China, it is built on the premise that global AI is an inherently competitive “race,” and China is clearly the US’s main competitor. A few days after the US AI Action Plan was released, China released its “Global AI Governance Action Plan.” It builds on the 2023 Global AI Governance Initiative, which seeks to portray China as a leader in the global pursuit of AI for the good of humanity and a “champion of the Global South.” The Global AI Governance Action Plan contains fewer specifics than the US AI Action Plan, but calls for “strengthening international cooperation on AI capacity building,” including supporting developing countries to design, implement, and govern AI. It also calls for building a global AI governance system through the UN and again shows the key balance of economic development and social stability: while it acknowledges the need to “jointly push for innovation breakthroughs,” it also advocates for strengthening “policy and regulatory coordination” and building an international AI safety governance framework. The US, on the other hand, is determined to cut “bureaucratic red tape and onerous regulation” while working primarily with its allies. However, recent events have called into question the feasibility of this approach. Notably, despite calls to increase export controls—and mobilise allies to do the same—the US announced shortly before the release of the AI Action Plan that it would again permit Nvidia to export H20 chips to China after previously restricting them. This was part of a trade deal in which China agreed to lighten restrictions on rare earth exports. Though seemingly counterintuitive, this deal reflects a fundamental reality of the modern AI ecosystem: the resources and supply chains enabling it are inextricably intertwined. China mines roughly 70 percent and processes 90 percent of rare earth minerals. US-based Nvidia has 92 percent of the GPU market. Both of these are critical for progress in AI. Both countries are attempting infrastructure decoupling, but the US may have to compromise on its fundamentalism to maintain access to the critical resources it needs. Meanwhile, its advanced chip dominance in turn provides leverage over China. Will this lead to broader cooperation? Geopolitical issues are likely too entrenched for each country to join hands and promote harmonious global AI governance. However, each country’s primary goal is the same: to benefit the “people.” China is claiming that its definition of “the people” is the global community—though this comes with the noted exclusion of domestic critics, including the Uighurs of Xinjiang, who are subject to AI-enabled surveillance and detention. The US’s definition of “the people” has changed from administration to administration, but currently seems to include the American people and potentially allied nations. It’s doubtful that the current administration will agree to substantive global AI governance discussions. But through its Action Plan, China has thrown down the gauntlet—engage with global AI governance or be an obstructionist. What the US will choose remains to be seen. Emmie Hine is a Research Associate at the Yale Digital Ethics Center and a PhD candidate in Law, Science, and Technology at the University of Bologna and KU Leuven. She researches the ethics and governance of emerging technologies, including AI. You can also find her byline in her weekly tech-focused newsletter, the Ethical Reckoner. Emmie holds degrees from Williams College and the University of Oxford, and previously worked as a software engineer. She’s on Bluesky here and X here. This article is published under a Creative Commons License and may be republished with attribution.

The BRICS meeting in Rio on July 6th and 7th gives a snapshot of the great power competition between China and the United States in different regions around the world, including Latin America. China has become the largest trading partner for many countries in Latin America, investing heavily in infrastructure and forging political alliances that further its strategic objectives. For its part, the Trump Administration of the United States issued the statement that those participating countries will face increased tariffs. The statement was the continuation of exercise and assertion of its authority for the past and present century. The positioning of various BRICS members and participating countries is particularly telling of what the great power competition means in the region and also globally. Brazil’s Lula hosted the meeting aiming to showcase its foreign policy leadership, not necessarily antagonizing the West. Russia is still going through the war in Ukraine, and Putin attended only online. India’s Modi was present as well as Ramaphosa from South Africa. No show of Xi Jinping was notable, although Premier Li Qiang was attending. Besides the BRICS core, other countries also showed promotion of their interests. Iran, for one, joined the group in 2024 and sent a ministerial level delegation to rebuke recent strikes on Iran. As the United States appears to be pulling back from its traditional leadership role in the world, China is seizing the opportunity to expand its influence and reshape global dynamics. Through a combination of state-driven development policies and active international engagement, Beijing has positioned itself as a major player in the Global South, extending its reach beyond Asia to regions such as Africa and Latin America. China’s increasing presence in the region has been mainly driven by the Belt and Road Initiative (BRI) and a surge in trade volumes, marking a major shift in the region’s economic landscape. Many experts point to China’s use of “infrastructure diplomacy”—financing ambitious, strategic infrastructure projects across the region—as a key factor in this rise. The numbers tell a compelling story. Trade data from the World Bank (Figure 1) shows that in the past ten years, China has overtaken the United States as the leading trading partner for much of the region, upending a dynamic that had held steady since the early 2000s. Beyond trade, China’s influence deepens through the 22 countries in Latin America and the Caribbean that have joined the Belt and Road Initiative. Chinese loans have poured in, funding major energy, infrastructure, and development projects that have reshaped local economies. China’s push isn’t just economic—it’s political too. Beijing has taken steps to strengthen cultural ties, increase academic exchanges and boost tourism in Latin America, including waiving visa requirements for travelers from some countries. This multi-faceted approach highlights China’s pragmatic mix of economic self-interest and strategic diplomacy as it works to secure resources, expand markets, and bolster its global standing. On the other hand, the United States has long been a strategic ally and key trading partner for Latin America. Agencies like USAID have funneled millions of dollars into economic and military initiatives across the region. With the recent changes in the aid policy, immigration policy, and tariff policy, Washington’s recalibration of its foreign policy are transforming the geopolitical balance in Latin America and the Caribbean. As both powers deploy their strategies — from deepening economic ties to defending national interests — the decisions of Latin American states remain critical in shaping their alignments with global powers. The ultimate outcome is still up in the air, but one thing is clear: power in the region is actively being renegotiated. The diverging approaches from China and the US have set the stage for a broader reconfiguration of power in Latin America and the Caribbean. Yet, it’s essential to recognize that each country’s internal decisions and policies also play a critical role in shaping this shifting landscape. Colombia provides a case in point. Historically, it has maintained close diplomatic ties with the United States while keeping China at arm’s length. Unlike countries like Brazil, Argentina, and Peru, Colombia has received relatively little Chinese infrastructure investment. However, with the election of president Gustavo Petro—the first left-wing president in the country’s history—Colombia has taken decisive steps to strengthen its relationship with China, presenting new challenges for the United States to maintain its strategic foothold in the country. We observe – both on political and economic dimensions – that the changes in China’s strategy, coupled with Colombia’s domestic policies, have reduced the country’s dependence on the US while increasing its desire to integrate with China. Politically, Colombia and the United States have long enjoyed a strong diplomatic relationship, as reflected in their shared memberships in international organizations, high-level dialogues, and multiple bilateral agreements. However, diplomatic tensions have emerged in recent years. Disputes between the two leaders, the change of course of USAID, and a significant drop in new bilateral agreements over the past four years have contributed to a shift in this traditionally stable partnership. Against this backdrop, diplomatic ties between China and Colombia have strengthened. In 2023 alone, both countries signed 12 cooperation agreements in trade, technology, and economic development, upgraded their relationship to a strategic partnership, and Colombia’s entry into the Belt and Road Initiative during recent China – CELAC Forum in May. Colombia also joined the BRICS New Development Bank a few weeks after that Forum. Economically, the US has traditionally been Colombia’s largest trading partner, backed by a free trade agreement and significant investment. Yet, in recent years, the share of US trade has steadily declined, while China’s footprint has grown (see figure 1). Although there’s no formal trade agreement, ties have strengthened during the current administration, including the opening of a Buenaventura-Shanghai trade route in 2025. Additionally, China’s “infrastructural diplomacy” has significantly grown: over 100 Chinese companies now operate in Colombia, and major infrastructure projects like Bogotá’s Metro Line 1 and the Regiotram are underway, along with investments in mobility, technology, and health. Latin America, and Colombia in particular, finds itself at the center of a geopolitical tug-of-war with China’s calculated investments and the US’s shifting policies. While Beijing leverages trade, infrastructure, and cultural diplomacy to expand its influence, Washington’s recalibration of its foreign policy leaves room for new alliances and opportunities. Our analysis shows that power reconfiguration is not merely a product of external rivalry. It is driven by the choices each Latin American nation makes. As Colombia’s case demonstrates, the region’s destiny hinges not just on global superpowers, but on its own internal political decisions and developments. The coming years will test how Latin America navigates these shifting currents. Disclaimer This article was made possible in part by a grant from the Carnegie Corporation of New York (G-PS-24-62004, Small State Statecraft and Realignment). The statements made and views expressed are solely the responsibility of the authors. Figure 1: China vs. US Import and Export TrendsDrawn by the authors using data from the World Bank.  

On March 4, the U.S. Department of Justice charged ten Chinese nationals with carrying out massive hacks against government agencies, media outlets, and dissidents in the United States and around the world. They allegedly acted on behalf of the Chinese company i-Soon, under contract from the Beijing government. Two officials from China’s Ministry of Public Security (MPS) were also indicted, identified as the ones “directing the attacks.” According to documents held by the U.S. justice system, China’s domestic intelligence services (MPS) and foreign intelligence (Ministry of State Security, MSS) relied on a vast network of private companies and domestic contractors to hack and steal information, thereby masking the Chinese government’s direct involvement. In some cases, the MPS and MSS paid private hackers to target specific victims. In many others, the attacks were speculative: hackers identified vulnerable computers, breached them, and extracted information that was later sold — either directly or indirectly — to the Chinese government. The Growth of Chinese Cyberespionage and Its Main Areas of Operation This is not an isolated case. Over the past decade, the People’s Republic of China’s (PRC) hacking program has expanded rapidly. In 2023, then-FBI Director Christopher Wray stated that it was larger than that of all other world powers combined. This increase in power and sophistication has led to success in three key areas: political interference, sabotage of critical infrastructure, and large-scale intellectual property theft. Beijing integrates computer networks, electronic warfare, economic, diplomatic, legal, military, intelligence, psychological, and military deception resources, along with security operations, to weaken states, make them economically dependent on China, and more receptive to a “new authoritarian world order with Chinese characteristics.” For this reason, unlike traditional interpretations, Chinese state-sponsored hacking should be understood within a broader context — where control over technology, strategic infrastructure, and global supply chains is part of “trans-military” and “non-military” warfare operations, as described by two People’s Liberation Army (PLA) colonels in the 1999 book “Unrestricted Warfare”. This approach is known as liminal warfare — an escalating conflict in which the spectrum of competition and confrontation with the West is so wide that the battlefield is, quite literally, everywhere. Cyberespionage as a Tool of Electronic Warfare In electronic warfare, hacking is used for sabotage during times of crisis or conflict. These actions are led by the People’s Liberation Army (PLA), the armed wing of the Chinese Communist Party. In 2023, it was discovered that a hacker group linked to the PLA, known as “Volt Typhoon”, had infiltrated a wide range of critical infrastructure in the U.S. for years, including ports, factories, and water treatment plants — both on the mainland and in strategic locations like Guam. “Volt Typhoon is a military operation with political and potentially military strategic purposes,” explained Ciaran Martin, former director of the UK’s cybersecurity agency. Led by the PLA’s cyber unit, the operation involved installing readiness capabilities — “digital traps,” as some call them — within critical U.S. infrastructure. In addition to a sustained attack in 2023 on a power company in Massachusetts, which aimed to extract sensitive data about its operational technology (OT) infrastructure, “Volt Typhoon” gained notoriety for multiple attacks on telecommunications systems in the U.S. and other critical infrastructures globally. One of its subunits, “Voltzite”, targeted the Littleton Electric and Water Departments, prompting the FBI and cybersecurity firm Dragos to respond jointly and publish a detailed report on the attack and its mitigation. Intellectual Property Theft Through Cyberespionage The most damaging channel for intellectual property theft is cyberespionage. These intrusions allow Chinese companies — sometimes with direct support from the Communist Party or the state — to access information on operations, projects, and technology from foreign firms. China has used state-backed and coordinated cyberespionage campaigns to steal information from companies in strategic sectors such as oil, energy, steel, and aviation. These actions serve both to acquire science and technology and to gather intelligence useful for future attacks on military, government, or technical systems. In the United States, there have been numerous precedents: • In 2014, five PLA hackers were indicted for economic espionage.• In 2017, three hackers linked to the Chinese firm Boyusec were charged with stealing confidential business information.• In 2018, two Chinese nationals were indicted for intellectual property theft.• In 2020, two hackers connected to the MSS were charged with targeting COVID-19 research. Among these, the 2018 indictment stands out as part of a broader U.S. effort to raise awareness about Chinese cyberespionage. On that occasion, Chinese hackers carried out a campaign known as “Cloud Hopper”, which involved a supply chain attack on service providers like Hewlett Packard and IBM. The defendants worked for Huaying Haitai and collaborated with the Tianjin State Security Bureau of the MSS. In 2017, the U.S. Commission on the Theft of American Intellectual Property estimated that such crimes cost the U.S. economy up to $600 billion annually — a figure comparable to the Pentagon’s defense budget and greater than the combined profits of the 50 largest companies in the Fortune 500. Beyond the United States: The Global Impact of Chinese Cyberespionage In June 2024, Dutch military intelligence (MIVD) warned that Chinese cyberespionage was broader than previously believed, affecting Western governments and defense companies. A 2023 cyberattack on the Dutch Ministry of Defense reportedly affected at least 20,000 people within a few months. In 2018, the Czech Republic’s National Cyber and Information Security Agency (NUKIB) issued a warning about risks linked to China. Since then, the country has strengthened its capabilities and controls against Beijing and has worked on mechanisms to counter foreign information manipulation. According to U.S. prosecutors, dozens of European parliamentarians have been targeted by Chinese attacks. In March 2024, the U.S. Department of Justice indicted hackers linked to the MSS for attacking “all EU members” of the Inter-Parliamentary Alliance on China (IPAC), a coalition critical of Beijing. In 2021, the hackers sent over a thousand emails to around 400 accounts linked to IPAC, attempting to spy on their internet activity and devices. In addition, ASML, the Dutch leader in semiconductor lithography, suffers “thousands of security incidents per year,” including several successful infiltration attempts by Chinese actors. Research centers like Imec (Belgium) are also frequent targets. Belgium has expelled Chinese researchers suspected of espionage. The European Union has reinforced security and identified advanced semiconductors as one of four critical technologies requiring risk assessments and enhanced protection. Notably, APT41 is one of the most active and sophisticated Chinese cyberespionage groups, based in the PRC and linked to the MSS. According to Google’s Threat Intelligence Group, APT41 combines state espionage with ransomware attacks — malicious programs that encrypt files and demand financial ransom to restore them — making attribution more difficult. Unlike other PLA-aligned groups whose operations are region-specific, APT41 acts globally, attacking strategic sectors in the U.S., Europe, Latin America, and the Caribbean. It also carries out financially motivated operations, particularly in the gaming industry. Mandiant, a global cybersecurity leader, highlights APT41’s technical capabilities: it frequently exploits zero-day and n-day vulnerabilities and uses techniques like phishing, social engineering, and SQL injections. Since 2020, APT41 has conducted large-scale campaigns against over 75 companies in more than 20 countries. It is responsible for compromising supply chains, such as in the “ShadowHammer” campaign targeting ASUS, which affected over 50,000 systems in 2018. APT41 is also linked to the use of “MESSAGETAP” malware in telecommunications networks. The Role of Chinese Universities in Cyberespionage Chinese universities also collaborate with the PLA and MSS in state-sponsored cyberespionage operations. Shanghai Jiao Tong University works directly with the Chinese military on such operations. Zhejiang University and the Harbin Institute of Technology are key centers for recruiting hackers. Xidian University offers students hands-on experience at provincial MSS offices and previously maintained ties with the Third Department of the PLA’s General Staff before its reorganization in 2015 into the Network Systems Department. One of its graduate programs is co-directed with the Guangdong Office of the Chinese Information Technology Security Evaluation Center (ITSEC), an MSS-run office that leads an active team of contractor hackers. Southeast University also maintains links with security services and co-manages the “Purple Mountain Lab” with the PLA’s Strategic Support Force. There, researchers collaborate on “critical strategic requirements,” operating systems, and interdisciplinary cybersecurity studies. The university also receives funding from the PLA and MSS to develop China’s cyber capabilities. The Cybersecurity undergraduate program at Shanghai Jiao Tong University (SJTU) is taught at a PLA information engineering base. Within this program, SJTU claims to work on “network and information systems testing and evaluation, security testing for connected smart networks, APT attack and defense testing, and key technologies for cyber ranges.” Universities associated with the MSS for talent recruitment include the University of Science and Technology of China, Shanghai Jiao Tong University, Xi’an Jiao Tong University, Beijing Institute of Technology, Nanjing University, and the Harbin Institute of Technology. In addition, some cybersecurity firms — such as Beijing TopSec — collaborate with the PLA in hacking campaigns, operator training, and developing future hackers. This article was originally published by Agenda Digitale and later by Expediente Abierto, who granted us permission for its translation and republication.

ABSTRACT  This paper examines de-dollarization as a geopolitical industrial policy within the BRICS framework and its implications for Cuba, Venezuela, and Argentina. De-dollarization, a process aimed at reducing reliance on the US dollar, has gained momentum among BRICS nations as a response to economic sanctions, monetary sovereignty concerns, and external financial shocks, particularly following the 2008 global financial crisis. For Cuba and Venezuela, de-dollarization is necessary due to US sanctions, pushing them toward alternative  financial  mechanisms  through  BRICS  partnerships. Cuba’s  possible  de-dollarization  follows  increased ties with Russia, China, and Iran. Regarding Venezuela, despite its partial dollarization, Caracas seeks  to  strengthen  non-dollar  transactions  through  oil  trade. In  contrast,  under  President  Javier  Milei,  Argentina  has  rejected  BRICS  and  continues  to  debate  dollarization,  reflecting  the  country’s  historical  and economic ties to the US dollar. The study highlights that de-dollarization is a State-led, multilateral process influenced by external economic conditions and geopolitical alignments. While Cuba and Venezuela actively integrate with BRICS to reduce dollar dependence, Argentina’s approach remains uncertain, shaped by ideological and financial considerations. Keywords: De-dollarization, BRICS, Cuba, Venezuela, Argentina INTRODUCTION De-dollarization is almost a synonym of BRICS. The reduction  of  US  dollar  dominance  and  the  consequential dependence on it represent critical stakes for BRICS countries. Nonetheless, there are nuances and differences amongst BRICS members on monetary policies. Since the first summits (2009–2010), BRICS  asserted  the  Global  South’s  need  to  prioritize  trade  in  domestic  currency  and  refrain  from  US  dollar  pegging. For  initial  members  like  China  and  Russia,  as  well  as  newly  associated  countries  such  as  Iran  and  Cuba,  Western  sanctions  are  the  main  driver  for  de-dollarization. Instead,  for  Brazil,  India, and the majority of most recent BRICS partners  (primarily  from  Africa  and  Southeast  Asia),  de-dollarization  means  enhancing  their  monetary sovereignty,  fostering  domestic  currencies’  value,  and  avoiding  depending  on  US  institutions:  Treasury and Federal Reserve. De-dollarization pertains to  monetary  and  public  policies. Therefore,  it  is  a  state-led process. For this reason, it might be considered an industrial policy. It is necessary to outline that this article adopts the term geopolitical industrial  policy  for  a  State-led  economic  strategy  that,  unlike  inward-oriented  monetary  or  financial  policies, is deeply intertwined with the outward-looking dimension of foreign policy. Hence,  this  work  examines  de-dollarization  as  a  geopolitical  industrial  policy  within  the  BRICS  framework  and  its  implications  for  Cuba,  Venezuela,  and  Argentina. De-dollarization,  a  process  aimed  at  reducing  reliance  on  the  US  dollar,  has  gained momentum  among  BRICS  nations  as  a  response  to economic sanctions, monetary sovereignty concerns,  and  external  financial  shocks,  particularly  following the 2008 global financial crisis. For Cuba and  Venezuela,  de-dollarization  is  necessary  due  to  US  sanctions,  pushing  them  toward  alternative  financial  mechanisms  through  BRICS  partnerships. Cuba’s  possible  de-dollarization  follows  increased  ties  with  Russia,  China,  and  Iran. Regarding  Venezuela, despite its partial dollarization, Caracas seeks to  strengthen  non-dollar  transactions  through  oil  trade. In  contrast,  under  President  Javier  Milei,  Argentina has rejected BRICS and continues to debate dollarization, reflecting the country’s historical and economic ties to the US dollar. The study highlights that de-dollarization is a State-led, multilateral process  influenced  by  external  economic  conditions  and geopolitical alignments. While Cuba and Venezuela actively integrate with BRICS to reduce dollar dependence,  Argentina’s  approach  remains  uncertain,  shaped  by  ideological  and  financial  considerations. It is undebatable that there are differences between usual industrial policies and de-dollarization. Indus-trial policies look inward, are fashioned upon domes-tic  matters,  and  contradict, court,  multilateral  efforts. De-dollarization  is  a  geopolitical  industrial  policy that looks outward, focusing on the role of a given country in the world economy. Without multilateralism, a State pursuing de-dollarization would quickly become a pariah. As a geopolitical industrial policy,  de-dollarization  owes  its  rationale  to  external  shocks. It  is  safe  to  define  de-dollarization  as  exogenously  motivated. The  2008  global  financial  crisis (GFC) represented the critical external shock for  BRICS  members  to  escalate  their  de-dollarization objectives: “[E]specially  since  the  2008  global  financial  crisis,  central banks of many countries have been trying to diversify their portfolios to shift away from the US dollar through liquidating holdings of US Treasuries and increasing other assets including the euro, yen, renminbi and gold.” (Li, 2023, p. 9).  The 21st century wrought incentives to de-dollarization that finally sparked because of the GFC. However,  the  mainstream  doubts  surrounding  de-dollarization involve its feasibility. There are no tools to objectively  measure  the  status  of  de-dollarization  or its future outcomes. Notwithstanding limitations, de-dollarization  is  increasingly  attracting  Global  South economies. Specifically looking at Latin America,  this  work  outlines  how  de-dollarization  becomes  an  obligation  for  sanctioned  countries:  Cuba  and  Venezuela. The  two  ALBA  governments  mingled  with  BRICS  for  a  long  time,1  with  Havana  joining the forum in association and Venezuela almost on the same route, stopped by the Brazilian veto in the  Kazan  summit. Cuban  and  Venezuelan  de-dollarization finds in BRICS a multilateral opportunity.  The third country examined is Argentina since the government  of  Javier  Milei  refused  to  enter  BRICS  and  continuously  flirted  with  dollarizing  the  economy. From President Menem’s pegging to the US dollar (uno a uno) to the 2001 Corralito, Argentina’s recent economic history inevitably rests on currency issues (IMF, 2003). Unlike Venezuela, and on the contrary of Cuba (which is not part of the IMF), Argentina’s economic policies intertwine with Bretton Woods  institutions. That  might  be  the  reason  why  neoliberal Argentinian economists found in dollarization  a  solution  for  Buenos  Aires  (Cachanosky  et  al., 2023).  1. Force Majeure De-Dollarization for Cuba and Venezuela  Since  1999,  when  Fidel  Castro  and  Hugo  Chávez  coincided, de-dollarization meant an industrial foreign policy to antagonize US hegemony. In Cuba, de-dollarization  is  a  more  difficult  process  than  usual  assumptions  and  certainly  more  challenging  than  in  Venezuela. 2004  marked  the  year  when  the  US  dollar  was  officially  prohibited  on  the Caribbean Island, to reverse the dual currency  system  implemented  since  the  Special  Period  (Herrera  &  Nakatani,  2004). The  extra-territoriality  of  US  sanctions,  affecting  in  their  secondary effect  Cuba’s  trade,  led  Havana  to  a  de-dollarization fashioned upon the path that Deligöz (2024) identified  for  China  and  Russia. Besides  realpolitik  and  geopolitical  strategies,  Cuba’s  association  with  BRICS,  occurred  in  October  2024,  is  the  la-test  effort  to  de-dollarize. Venezuela’s  economic  crises and COVID-19 pushed Cuba into continuous indebtedness to survive, with US dollars reallowed but  still  at  limited  provision  due  to  Washington’s  restrictions  (Luis,  2020). To  give  account  of  its  urgencies,  in  a  few  months,  Havana  moved  from  apparent dollarization to initiatives for de-dollarization, thanks to BRICS. Over the summer, Primer Minister Manuel Marrero enabled USD payments in the  tourist  sector  (Gámez  Torres,  2024)  to  tackle  the balance of payments deficit with liquidity. For  a  country  obliged  to  rapidly  change  industrial  policies,  the  BRICS  opportunity  could  not  be  mis-sed. Cuba’s  reliance  on  Russia,  China,  and  Iran  may  materialize   a   complete   de-dollarization   that   can   favor  BRICS  projects  and  escape  US  sanctions. Of  course, the evident permanence of the bloqueo, regardless  of  who  runs  the  White  House,  is  the  main  driver for Cuba’s de-dollarization. A similar but quite nuanced situation applies to Venezuela as well. From the Bolivarian era inaugurated by President Chávez, de-dollarization  entangled  foreign  policy  objectives  even before US sanctions. The Sucre digital currency was  created  by  the  governments  of  Venezuela  and  Ecuador  as  the  main  ALBA  initiative  to  de-dollarize  commercial  transactions  among  Bolivarian  nations  (Benzi et al., 2016). ALBA-promoted Sucre was analogous to BRICS’ favoring of blockchains and digital currencies, limiting the USD to a reference value for the  bloc’s  transactions  (Mayer,  2024). US  sanctions  on Venezuela’s oil production, sparked under the first Trump Administration, meant a significant remotion of USD-denominated transactions for Caracas. Considering  ALBA’s  slow  progress  and  the  infeasibility  of fully adopting the Sucre, President Maduro had to look at BRICS for solutions. Despite  not  having  diplomatic  relations  with  Washington,  Venezuela  is  still  an  IMF  member. Ladasic points  out  that  “[a]s  Venezuela  joined  the  pack  of  countries  trading  oil  outside  of  USD  and  has  instead priced it in Chinese yuan, BRICS together with Venezuela  already  have  16%  needed  for  IMF  veto  power to use in a crisis” (2017, p. 100). The rentier characterization of the Venezuelan economy and its dependency  on  oil  exports  make  de-dollarization  a necessity. As per Cuba, unilateral policies are not enough. Venezuela’s  outcry  merged  with  inflation,  the  devaluation  of  the  bolívar,  and  a  paralysis  of  the  Venezuelan  Central  Bank  (BCV)  that  put  total  dollarization on the industrial-public policies’ table (Briceño  et  al.,  2019). Although  the  country  is  still  under  a  sort  of  de  facto  dollarization,  Venezuela’s  economic  resurrection  should  occur  together  with  a  de-dollarization  strategy. Failure  to  enter  BRICS  in the Kazan summit provides a temporary brake to Venezuela’s  de-dollarization,  but  the  prolific  trade  with China, Russia, Iran, and Türkiye will, in all cases, align Venezuela with BRICS policies. 3. Argentina: De-Dollarizing a Passion Economists  were  surely  interested  in  Javier  Milei’s  dollarization  claims. Less  than  a  year  into  his  government,  dollarization  seems  impossible  to  the  libertarian  president. Milei’s  negative  to  BRICS  demonstrates  that  de-dollarization  is  currently  not  considerable  for  Casa  Rosada. Nevertheless,  it  is  relevant to outline that Argentinian academia questioned  the  role  of  the  USD  and  studied  economic  policies  involving  de-dollarization. Corso  and  Sangiácomo (2023), in affiliation with the Central Bank of  Argentina  (BCRA),  argued  that  de-dollarization  might  help  in  relieving  the  extreme  inflation  saw  under  Alberto  Fernández’s  ruling. Other  authors  implied  that  the  Kirchners’  limitations  on  USD  access would lead to a gradual de-dollarization of the economy,  but  with  constraints  particularly  from  a  USD dominated housing market across Latin America  (Luzzi,  2013). If  under  the  Kirchners,  and  with  support of South American left-leaning geopolitics, de-dollarization  could  really  offer  a  pathway  for  the Argentine economy, with Milei that is barely an option. The  Argentine  relation  with  the  USD  does not hold a clear ideological cleavage. Argentinians’ passion for the dollar, as stressed by Bercovich and Rebossio (2013), embraced diverse political figures such as Perón, Aníbal Fernández (a prominent Kirchnerist politician), and Martínez de Hoz. The peso’s continuous  instability  legitimized  the  widespread  informal adoption of the USD, with first insight fore-seeable in the currency devaluation subsequent to the Great Depression (Díaz Alejandro, 1970). There is also a nationalistic meaning behind the peso, whose  national  heroes  imprinted,  from  Belgrano  to  Evita (Moreno Barreneche, 2023), portray a sentimental attachment to the banknotes that Argentinians do not want to erase. In sum, Argentina’s de-dollarization is as difficult as dollarization. Milei’s obsession for US hegemony inserts de-dollarization in a faraway scenario. Moreover,  Donald  Trump’s  victory,  who  promised  high tariffs to countries that unpeg from the USD (Butts,  2024),  constitutes  a  natural  barrier  to  de-dollarization. Its political viability might depend on an eventual Peronist succession to Milei. Argentina’s financial closeness  to  China,  and  a  possible  resume  of  BRICS  talks,  could  indicate  de-dollarization  as  a  future  last  resort. In this sense, de-dollarization within the BRICS framework might help Argentina in solving structural issues: Chronic external debt and dependency on Bretton Woods institutions. CONCLUSIONS De-dollarization is State-led and can be considered a  geopolitical  industrial  policy. Cuba,  Venezuela,  and  Argentina  show  that  de-dollarization  depends  on  geopolitical  calculus  and  economic  considerations. The incentives may be different, ranging from US sanctions to devaluation of the national currency. However,  unlike  dollarization,  de-dollarization  cannot  be  pursued  unilaterally. The  rise  of  BRICS  motivates  Global  South  countries  to  de-dollarize  under its guarantees. For Cuba and Venezuela, the association  with  BRICS  and  the  interdependence  with other sanctioned economies like Russia, China, and Iran, make de-dollarization an opportunity. Argentina’s  relation  with  the  USD  follows  its  turbulent  economic  history. Simultaneously,  there  is  passion  for  dollars and nationalism toward the peso banknotes. In this  context,  even  Milei  showed  that  dollarization  is  in  no way easier that de-dollarization. The currency issues affecting Argentina might not be resolved by neither of the two policies, but a future BRICS collaboration could bring de-dollarization again into the political debate. NOTES1  ALBA  references  the  Alianza  Bolivariana  para  los  Pueblos  de  Nuestra  América,  a  regional  organization  founded  by  Cuba  and  Venezuela,  including Bolivia, Honduras, Nicaragua, and several Caribbean islands. It was created in 2004 under the auspices of Hugo Chávez.REFERENCESBenzi,  D.,  Guayasamín,  T.,  &  Vergara,  M.  (2016). ¿Hacia  una  Nueva   Arquitectura   Financiera   Regional?   Problemas   y  perspectivas  de  la  cooperación  monetaria  en  el  AL-BA-TCP. Revista Iberoamericana de Estudios de Desarrollo, 5(1), 32–61. https://doi.org/10.26754/ojs_ried/ijds.193. Bercovich, A., & Rebossio, A. (2013). Estoy verde: Dólar, una pasión argentina. Aguilar.Butts, D. (2024, September 9). Trump’s vow of 100% tariffs on nations that snub the dollar is a lose-lose for China and U.S., economist says. CNBC. https://www.cnbc.com/2024/09/09/economist-calls-trumps-threat-to-tariff-countries-that-shun-the-dollar-a-lose-lose.html. Cachanosky, N., Ocampo, E., & Salter, A. W. (2023). Les-sons from Dollarization in Latin America. Free Market Institute  Research  Paper  No.  4318258,  AIER  Sound  Money  Project  Working  Paper  No.  2024-01.  https://doi.org/10.2139/ssrn.4318258. Corso, E. A., & Sangiácomo, M. (2023). Financial De-dollarization in Argentina: When the wind always blows from the East. BCRA Economic Research Working Paper No. 106. https://www.econstor.eu/handle/10419/297801.Deligöz, H. (2024). The Exorbitant Privilege of US Extra-territorial  Sanctions.  İnsan  ve  Toplum,  14(3),  29–52.  https://dergipark.org.tr/en/pub/insanvetoplum/is-sue/86942/1543025. Díaz Alejandro, C. F. (1970). Essays on the Economic His-tory of the Argentine Republic. Yale University Press.Gámez  Torres,  N.  (2024,  July  18).  Cuba  moves  to  ‘partially’  dollarize  economy  as  government  struggles  to  make  payments.  Miami  Herald.  https://www.mia-miherald.com/news/nation-world/world/americas/cuba/article290210784.html. Herrera,  R.,  &  Nakatani,  P.  (2004).  De-Dollarizing  Cuba.  International  Journal  of  Political  Economy,  34(4),  84–95. https://www.jstor.org/stable/40470915. Hurtado  Briceño,  A.  J.,  Zerpa  de  Hurtado,  S.,  &  Mora  Mora,  J.  U.  (2019).  Dollarization  or  Monetary  Independence?  Evidence  from  Venezuela.  Asian  Journal  of  Latin  American  Studies,  32(4),  53–71.  https://doi.org/10.22945/ajlas.2019.32.4.53. IMF. (2003, October 8). Lessons from the Crisis in Argen-tina. Ladasic,  I.  K.  (2017).  De-Dollarization  of  Oil  and  Gas  Trade.  International  Multidisciplinary  Scientific  Geo-Conference,    17,    99–106.    https://doi.org/10.5593/sgem2017H/15. Li,  Y.  (2023).  Trends,  Reasons  and  Prospects  of  De-Dollarization. South Centre Research Paper No. 181. https://www.econstor.eu/handle/10419/278680. Luis, L. R. (2020, October 7). Cuba: Dollar Crunch, Dollarization and Devaluation. Cuba Capacity Building Project. https://horizontecubano.law.columbia.edu/news/cuba-dollar-crunch-dollarization-and-deva-luation. Luzzi,  M.  (2013).  Economía  y  cultura  en  las  interpretaciones sobre los usos del dólar en la Argentina. In  A.  Kaufman  (Ed.),  Cultura  social  del  dólar  (pp.  11–19).  UBA  Sociales.  https://publicaciones.sociales.uba.ar/index.php/socialesendebate/article/view/3319.Mayer,  J.  (2024).  De-Dollarization:  The  Global  Payment  Infrastructure  and  Wholesale  Central  Bank  Digital  Currencies.  FMM  Working  Paper  No.  102.  https://www.econstor.eu/handle/10419/297865. Moreno  Barreneche,  S.  (2023).  El  dinero  como  soporte  material  de  la  disputa  por  el  sentido  de  la  nación:  Estudio  del  peso  argentino  desde  una  perspectiva  semiótica.  Estudios  Sociales:  Revista  Universitaria  Semestral,  64,  1–19.  https://doi.org/10.14409/es.2023.64.e0046. CONFLICT OF INTERESTThe  author  declares  that  there  are  no  conflicts  of  interest related to the article.ACKNOWLEDGMENTS Not applicable.FUNDING Not applicable.PREPRINT Not published.COPYRIGHT Copyright  is  held  by  the  authors,  who  grant  the  Revista  Política  Internacional  the  exclusive  rights  of  first  publication. Authors  may  enter  into  additional agreements for non-exclusive distribution of the  version  of  the  work  published  in  this  journal  (e.g.,  publication  in  an  institutional  repository,  on  a personal website, publication of a translation or as a book chapter), with the acknowledgment that it was first published in this journal. Regarding copyright, the journal does not charge any fee for the submission, processing, or publication of articles.

Abstract The paper considers the question of location in the development and governance of artificial intelligence in Africa. The discussion draws from ideas on locational advantage and the mix of factors that affect inequalities in AI development and how this influences the ability that countries have to shape AI norms, cultures and governance. It analyses policy documents and internet databases to highlight Africa’s place in AI development, the continent’s governance approach and the symbiotic relationship that explains the influence of advanced countries and tech corporations in the AI landscape. Based on this, it proposes the concept of the ‘politics of locationality’ to extend our understanding of how the power resident in AI systems is associated with their primary situatedness and how this reality, in turn, (re)produces imbalances and unequal opportunities for Africa in AI development and governance. It concludes with implications for Africa’s contribution to global AI cultures, design and governance at this time of pressing need for well-balanced AI policies. KeywordsAI governance, culture, ethics, inequality, locational advantage, politics of locationality Introduction The race for supremacy in AI development and governance has gathered pace among leading countries such as the US, European nations and China (Bradford, 2023). Many of the largest AI companies and models trace their origins to these countries and benefit from massive investments in computing, data and skills (Moorosi, 2024). By contrast, African countries mainly boast of AI startup ecosystems coordinated by technology hubs such as Kenya’s ‘Silicon Savannah’ (Eke et al., 2023) and African journalists have had to grapple with tools like ChatGPT, given the relevance of generative AI for media production (Gondwe, 2023). Although many of these experiences show Africa’s creative potential in the AI sector, as well as its growing market for AI technologies and tools (Okolo et al., 2023), there remain complex historical, cultural, political, legal, economic, labour and data factors that the continent faces in its drive towards AI development and governance. This mix of factors is a subject that scholars have considered to varying degrees (Muldoon and Wu, 2023; Png, 2022). I build on this by examining the question of locationality within that mix and how locationality helps to explain aspects of the symbiotic relationship between countries and platforms in shaping AI design and governance principles within a global context. My discussion sits at the intersection of critical media studies (Ott and Mack, 2014), digital geopolitics (Wong, 2021) and AI studies (Crawford, 2021), seeking to demonstrate why AI, understood as a political tool for constructing knowledge and generating maps of meaning, should be situated. This focus on situatedness further ties into studies on economic geography and locational advantage (Dunning, 1998; Iammarino and McCann, 2013), which explain why multinationals such as AI companies are more likely to choose certain locations as headquarters and how these locations are better placed to acquire greater leverage and power. It relates to the understanding that the location where a technology company is headquartered has considerable influence to shape that company’s decisions (Sargsyan, 2016), presupposing that advanced AI development confers locational advantage on the place where this development is primarily domiciled, making it a locus of power in shaping global AI ethics, norms and governance. This presupposition is what I interrogate by asking: How do issues around the location of AI development explain the impact that African countries have to shape AI design, cultures and governance in light of inequalities between advanced nations and the majority world? This question is crucial as it indicates which countries are better positioned as key definers of AI at a time when the design, norms, ethics, principles and governance of the technology are being concretised. It also leads to my argument on what I call the ‘politics of locationality’ as a concept to advance our understanding of how the primary situatedness of leading AI companies in advanced locations (such as the Global North) reveals inequalities in AI development and governance in the majority world. In this paper, advanced locations refer to the US, Europe and China – following Bradford’s (2023) description of them as the three digital empires, having the capacity not only to define the digital ecosystem in line with their cultures and priorities, but also shape the digital values of other countries that fall under their influence. The discussion that follows flows through a review of the literature, my methodology and findings, culminating in my argument on the politics of locationality. Locationality and the mix of factors in AI development Across the AI landscape, ethical principles are vital because they guide AI development, design and governance (Tidjon and Khomh, 2022). In Africa, some of the ethical and cultural values that researchers contend should guide AI development include ubuntu (communal humaneness), humanism and indigenous knowledge (Nayebare, 2019; Mhlambi and Tiribelli, 2023). Not enough of this has been realised, however, given that AI technologies on the continent remain a largely Western import (Eke et al., 2023). Nonetheless, home-grown African AI systems have been on the rise, evident in the creation of technology centres such as Kenya’s ‘Silicon Savannah’, Ethiopia’s ‘Sheba Valley’ and Nigeria’s ‘Yabacon Valley’ (Eke et al., 2023). The centres serve as ecosystems that provide leverage for AI startups that are supported by communities such as Data Science Nigeria and Deep Learning Indaba (Nuwer, 2024). Okolo et al. (2023) found that there are at least 102 of these startups in Africa. But they face challenges ranging from limited infrastructure for energy and computing-intensive technology and lack of expertise (Nuwer, 2024). These challenges presuppose that AI development on the continent is likely to lag behind global advancements – with implications for the influence that the continent has as a place from which AI norms, ethics and governance are shaped. The US remains a dominant player in shaping AI ethics and governance. Chan et al. (2021), for instance, observe that the US is predominant in the AI development space, in line with its economic and cultural dominance and China follows closely behind, pointing to a US-China tech war in digital geopolitics (Wong, 2021). China, in particular, has exported AI facial recognition and smart city technologies to Africa, serving as the continent’s largest foreign ICT investor (Lin, 2024), with increasing access to African datasets (Okolo et al., 2023) – a worrying trend. Also, investments in AI still concentrate in countries like the US and China, with only 1% of global venture capital flows into Africa (Moorosi, 2024). This means profits from AI development are typically reinvested in the leading countries, creating further disparities between them and the majority world (Chan et al., 2021). Although major tech companies (e.g. Google and Microsoft) have invested in Africa by establishing AI labs in African countries, Chan et al. (2021) note that it is common for staff working in these labs to be based in advanced countries. These major AI companies have also diversified their locational presence by outsourcing some operations through an international division of labour across Africa. But they tend to do so in exploitative ways such as hiring low-wage data annotators, highlighting the continent’s locational disadvantage (Ludec et al., 2023). What this points to is the relation of labour and capital, which further underscores Muldoon and Wu’s (2023) description of the global AI supply chain as colonial, since it is actualised through an international division of digital labour that extracts value from labour in the majority world for the benefit of Western technological companies. It also reinforces hegemonic knowledge production through Western values and knowledge that marginalise non-Western alternatives (Muldoon and Wu, 2023). This hegemonic knowledge production leads to questions on the role of the human in the machine, since AI is not based on abstract models, but is embedded in human agency and cultural values (Natale and Guzman, 2022). Additionally, this hegemonic knowledge production not only ties into discourses on power in critical media studies (Ott and Mack, 2014), but also on the relevance of AI for creative and cultural work (Lee, 2022) – as seen in the use of ChatGPT, which, African journalists say, is built on a poor and non-representative African corpus and perpetuates stereotypes of the continent (Gondwe, 2023). What the foregoing, therefore, indicates are the mix of historical, technological, economic, labour and cultural factors that underpin AI development and deployment in Africa’s locational context, with implications for governance outcomes. AI governance and digital regulation in Africa In terms of governance, state actors in Africa have drawn up policy and legal instruments at local levels and are only just taking part in global multistakeholder conversations. Examples of these multistakeholder fora include the AI Governance Alliance of the World Economic Forum (2025), where three African countries (Rwanda, South Africa and Kenya) are represented and the Global Partnership on Artificial Intelligence (2025), where Senegal is the only African country. It is unclear what influence African countries wield in multistakeholder bodies like these, but the indication is that inclusion is usually performative and exists for ‘virtue signalling and promotional purposes’ (Png, 2022: para. 17). At continental and regional levels, Plantinga et al. (2024) observe that African countries have increasingly turned towards AI strategies, but without considering how feasible these are to implement and their suitability to local contexts. There are also data localisation laws, which stipulate that data be stored and processed in the location where they are generated, with restrictions on data transfers across borders (Giovane et al., 2023). These laws, which point to the geographical relevance of data, have been introduced by at least 16 African countries, including Mauritius, Lesotho and Ivory Coast; they are also present in legal instruments on data protection, financial transactions, cybersecurity and telecommunications in several countries across the continent (CIPESA, 2022). Locationality is therefore the underlying principle in data localisation, underscoring the importance of data for AI control and governance. Sargsyan (2016: 2224), for instance, notes: ‘countries that host intermediaries’ data centres and offices have more opportunities to exercise influence over companies’ decisions and claim jurisdiction over data stored in their territory.’ What is crucial here is the presence of data centres which is vital for AI production, and African countries face challenges in relation to data, since the continent’s data centre landscape is nascent although growing (DCByte, 2023). When it comes to enforcement, only a few countries have been successful in imposing local laws and claiming jurisdiction over data stored in data centres, the most successful being the US (Sargsyan, 2016: 2231). This underscores the difficulties of AI policy implementation that African governments face, since, as CIPESA (2022) notes, there is barely any evidence that data localisation laws have been enforced in Africa. Where implementation has been more practicable is in the imposition of internet taxes on users in countries like Guinea, Benin, Uganda and Zimbabwe (Bergére, 2019). The infrastructure platforms through which governments enforce the taxes are local Internet Service Providers (ISPs) and telecommunication providers (Bergére, 2019). The same process finds expression in the imposition of internet bans in Africa – enforcement targets usage disruption and is done through local ISPs (Parks and Thompson, 2020). African countries ordinarily enjoy a positive balance of power in comparison with local ISPs. Also, the fact that major tech and AI companies can be banned from operating or delivering services in any African country points to the power that the state has to impose controls as a means of last resort. Users can also deploy collective action against AI companies (Fratini and Musiani, 2024). All these highlight the diverse ways in which power and governance are contested at local levels but say little about the place that Africa holds in contributing to and shaping AI ethics, governance and worldviews in the global sense. If, as Sargsyan (2016) notes, the location where a technology company is headquartered has significant powers to influence the decisions of the technology company, then what does Africa’s record in AI development imply for its ability to shape AI design, ethics, cultures and governance? This question is what I aim to answer in interrogating the relevance of the politics of locationality for AI in Africa. Method To this end, I draw from a methodological approach informed by textual analysis of AI policy documents and critical analysis of internet databases containing the location associated with AI companies/models. For the AI policy documents, I focussed on three domains. The first domain concerned the AI policies of African countries, where I selected the AI Strategies of four countries: Mauritius (Mauritius AI Strategy, 2018), Egypt (Egypt AI Strategy, 2021), Rwanda (Rwandan National AI Policy, 2022) and Nigeria [draft] (Nigerian AI Strategy, 2024), representing South, North, East and West Africa respectively. I also reviewed the African Union’s (2024) AI Strategy. Mauritius, Egypt and Rwanda are particularly relevant because they are the first three countries to introduce AI strategies in Africa. I chose AI strategies because they comprise one of the major emphases for policymakers on the continent (Maslej et al., 2024). I analyse the strategies based on themes, which I developed after reviewing and coding the texts. The second domain included the AI policies of leading global players in AI: the US, EU and China. Here, my analysis aimed to ascertain whether these policies mandate AI firms to develop their technologies in line with principles and standards upheld by the leading countries. The third domain included the policies of some of the largest AI companies/models: OpenAI and Gemini (for the US), Darktrace (for the UK) and UBTECH and Qwen (for China). My aim was to examine whether these AI companies/models subject their policies and practices to the countries or places where they are headquartered. For the database searches, I accessed and reviewed the following to ascertain the location of AI companies globally: Datamation (see Jungco, 2024), AI Directory (2024) and the Companies Market Capitalisation (2024). I also drew from sources such as Epoch AI (2024), the Emerging Technology Observatory’s (2024) AI Country Activity Tracker and the AI Index Report (Maslej et al., 2024). Overall, my goal was to assess Africa’s place in AI development in comparison with the leading countries in AI before considering the leverage that locationality affords to nations when it comes to AI design, ethics and governance – the underlying premise for the politics of locationality. Locational advantage and AI development in Africa While acknowledging the mix of factors that comprise AI development, my findings focus on locationality by considering the number of AI companies headquartered in Africa and elsewhere. Here, I found that most AI companies are headquartered in advanced countries, with many of the largest AI companies being in the US. These are companies that have significant capitalisation as listed on the Companies Market Cap to scale globally and invest in AI infrastructure and models that fundamentally impact societies. Jungco’s (2024) review shows that the overwhelming majority (80%) of these companies are based in the US (81 out of 101 companies). The closest to this was the UK, which had eight companies out of 101. India and Canada had three each; China and Germany had two each; and Israel and Hong Kong had one each. The same is true for data compiled by AI Directory (2024) and the Companies Market Capitalisation (2024). I compiled the three listings (Jungco, 2024; AI Directory and the Companies Market Cap) and removed duplicates. The combined list comprised 234 of the largest AI companies (see Table 1), ranging from Apple valued at $3.3 trillion to FR8Tech valued at $2.1 million. The outcome shows that at 73%, the US is the undoubted leader. It is followed by India (5.98%), the UK (5.56%), China (2.99%) and Canada (2.56%). There is no African country on the list.   There is a similar outcome for data on notable AI models, such as OpenAI’s ChatGPT, Alibaba’s Qwen and Meta’s Llama. Table 2 shows that, when isolating for AI models owned by organisations or entities that come from only one country, two-thirds (67.7%) of these notable AI models are traceable to the US. Again, no other country comes close; the closest countries are the UK, China, Canada and Germany. Maslej et al. (2024), in the AI Index, also show that foundation AI models, which are the basis for generative AI, are far more prevalent in the US. Again, there are no entries for Africa.   If Africa does not serve as a headquarter for the largest AI companies or models, what then can be said about AI startups? To answer this, I consulted the 2024 AI Country Activity Tracker, which presents data on the number of approved AI patents based on country locations, including patents for AI startups. The data, presented in Table 3, show that Africa is only ahead of the Middle East; even at this, Africa accounts for only 0.000006% or 157 of the 250,224 AI patents captured in the tracker. And only three African countries account for the continent’s figures: South Africa (139 patents), Morocco (17 patents) and Kenya (1 patent). Although this figure of 157 is slightly more than the 102 startups that Okolo et al. (2023) provided, the conclusions are similar: Africa has some way to go in making up for AI development in the future.   All these show that although Africa provides labour and data for AI companies globally and is poised to be a significant market as internet connectivity spreads (Access Partnership, 2024), it barely serves as a headquarter for AI companies, particularly the largest ones. This underscores Africa’s nascent standing in AI development and suggests that African countries will likely be constrained in their ability to shape wider AI ethics, cultures and governance. It perhaps explains why no African country has passed AI legislation into law (although some attempts are underway), with focus instead on AI task forces, national policies and strategies (Maslej et al., 2024). I turn next to the analysis of these strategies to examine the policy approach to AI development and governance in Africa. Analysing AI strategies in Africa My analysis of the AI strategies/policies of Mauritius, Egypt, Rwanda, Nigeria and the AU shows that they coalesce around four main themes: AI adoption for economic growth, AI partnership and collaboration, improving AI ethics and governance and strengthening local AI capacity. The focus on AI for economic growth is understandable, given that AI’s potential for economic transformation is a major attraction. This focus, which is present in all the strategies, highlights how AI can enhance economic development in various sectors (see in particular, the Mauritian Strategy). But more important, in the context of my discussion, are the inequalities in international cooperation and multistakeholder governance that the strategies point towards. To establish this, I refer to Pillar Two of the Nigerian AI Strategy, which underscores the need for partnership and collaboration among local and international stakeholders to leverage AI expertise and resources. The Rwandan AI Policy also recommends international partnerships and collaboration in AI development. But collaborations, especially with international private actors, highlight the need to evaluate where the major partners come from. These partners can collaborate with African countries and invest in AI, but they still operate primarily by rules set in their home countries, even if they also subscribe to laws in subsidiary countries (I establish this point in the next section). The exception would be collaboration within the continent, as can be found in the Egyptian and AU Strategies. The third theme, which focusses on ethics and governance, acknowledges the risks of AI and the need to mitigate them. It considers the importance of having AI principles such as fairness, transparency and accountability (see the Nigerian Strategy). The AU Strategy also mentions the need to guard against bias and risks to African values, indigenous knowledge and cultural heritage. This is connected to the AU’s Agenda 2063, which has a media and culture mandate to support inclusive and ethical AI through values such as ubuntu. But the plan to address these risks, for the most part, points to multistakeholder governance. The Egyptian Strategy, for instance, seeks to ‘actively contribute to global efforts and playing an active role in AI in different international fora’ (Section 5.1). Rwanda’s Policy also intends to ‘actively contribute to shaping responsible AI principles and practices in international platforms’ (Key Policy Recommendation 14). Likewise, the AU Strategy promotes a ‘multi-tiered governance approach’ (Section 2.4.1). They highlight a move towards greater inclusivity in international governance – hence the use of the word ‘actively’ in many of the strategies. My point, however, is that no matter how ‘actively’ African countries intend to influence AI cooperation, ethics and governance, there is a need to first reckon with Africa’s standing in AI development. Policymakers on the continent seem to have recognised this reality, and it is noticeable in their focus on strengthening local AI capacity. The AU Strategy, for instance, notes that Africa faces challenges around lack of computing platforms, limited data for training AI models and a scarce supply of AI skills, observing that, as of 2023, all the world’s supercomputers are located in only 30 countries. The Nigerian Strategy also recognises the need to build ‘affordable and localised infrastructure foundations and the compute capacity’ (Section 1.4.2) by making high-performance computing available. In Rwanda and Egypt, there is emphasis on AI training in schools. Evident here is a recognition that the continent lags behind in AI development, although it has a growing AI startup ecosystem and an unsaturated market for AI. It underscores the need to interrogate AI development, the locational advantages and disadvantages it represents and its links to governance inequalities. I argue that this should be a major consideration in Africa, based on what I describe as the politics of locationality. The politics of locationality Having established the foundational basis for my argument, I now define the politics of locationality as the way in which the power resident in Big Tech and AI systems is associated with their primary situatedness in places of advanced technological and AI development and how this underlying reality, in turn, (re)produces imbalances and unequal opportunities for AI development and governance faced by countries in the majority world, particularly Africa. The politics of locationality highlights the mix of advantages that countries such as the US, China and those in Europe where the largest AI companies are headquartered have. By contrast, African countries, despite advances that have been made in the AI startup ecosystem and state policy interventions, suffer from a range of disadvantages in locational, political, economic, cultural, data, labour and historical spheres. Addressing this mix of disadvantages is what the continent needs to become a location of influence from which the values, ethics and governance that circumscribe AI can be shaped. To expand on my argument, I refer to the AI policies of some of the largest companies and the guiding principles for AI that leading tech countries have drafted. Take OpenAI for instance. The OpenAI (2024) Terms of Use specify that dispute resolution between the company and users anywhere in the world would be coordinated by National Arbitration and Mediation, headquartered in New York. The law that governs the arbitration is the US Federal Arbitration Act. It adds that, ‘California law will govern these Terms’ and ‘claims arising out of or relating to these Terms will be brought exclusively in the federal or state courts of San Francisco, California’ (see Governing Law Section). This is not surprising, given that OpenAI is headquartered in San Francisco. Also, for Gemini (2024), owned by Google, the governing law of its User Agreement is the laws of New York, and the Agreement is seen as a ‘contract wholly entered into and wholly performed within the State of New York’ (see Governing Law Section). I found a similar pattern in the UK and China. In the UK, Darktrace (2024), an AI cybersecurity firm, notes that the governing law of its Master Services Agreement is the laws of England and Wales, except if the customer is located in the US, in which case the laws of California apply. In China, Section VII of the Terms of Use of UBTECH (2023), an AI robotics firm, states: ‘The establishment, effectiveness, performance, interpretation and dispute resolution of this [user] agreement are subject to the laws of the People’s Republic of China (excluding Hong Kong, Macau, Taiwan).’ The same goes for Alibaba’s Qwen (2023), which is covered by the Alibaba Cloud International Website Terms of Use, the governing law of which is that of Singapore, where Alibaba has its office in the Asia Pacific. Hence, I note that AI companies uphold the laws of the countries where they are headquartered, and they subject the governance of their technologies to these laws. AI companies have also come under sustained pressure to develop rules according to the values, ethics and standards of their host nations, thanks, in part, to the dictates of digital geopolitics and the competition for AI supremacy. The US, for instance, had the Biden Administration’s Executive Order on Safe, Secure and Trustworthy Artificial Intelligence (US Executive Order, 2023), through which it sought to promote an AI market that emphasises US innovation and leadership on AI (Section 2(h)) and the need to encourage international partners to support the voluntary commitments of US AI companies (Section 11(a)(ii)). But these voluntary commitments are underpinned by the need for US leadership and ‘to ensure that artificial intelligence is aligned with values shared by all Americans’ (proposed National AI Commission Act (2023), Section 3(g)(1)). The Biden Executive Order was replaced in January 2025 by the Trump AI Executive Order, aptly named, ‘Removing Barriers to American Leadership in Artificial Intelligence.’ All these underscore that a key intention of US policy is that the design of AI mirrors America’s geopolitical leadership, cultural norms, ethics and jurisprudence. A similar line of argument applies to Europe and China. In Europe, the first clause of the European Union (2024) AI Act says its purpose is to lay ‘a uniform legal framework’ for AI development and deployment ‘in accordance with Union values’ (Section 1) and it supports the ‘European human-centric approach to AI and being a global leader’ in AI development (Section 8). In China, there is the Generative AI Measures released in August 2023. The regulation includes a stipulation that generative AI technologies should not produce content that incite the ‘subversion of national sovereignty or the overturn of the socialist system’ (Henshall, 2023: para. 2). The same ethos is evident in China’s 2023 Management of Deep Synthesis of Internet information Services enacted by the Cyberspace Administration of China (2022). It mandates that in developing products such as AI, companies must abide by Chinese laws, correct political direction and value orientation. This was clearly evident in the January 2025 roll out of DeepSeek, a China-based AI assistant, which when asked about Tiananmen Square, refused to answer, saying it follows ‘ethical guidelines’ and added: ‘I respect the laws and cultural contexts in which I operate’ (author’s conversation with DeepSeek). It goes to show that expansive or restrictive design and deployment of AI technologies are subject to the value system prevalent in their originating countries. Hence, the politics of locationality sustains the different visions that leading tech countries have of AI in their battle for supremacy – visions that are not necessarily inclusive of African perspectives even though they find expression on the continent through pervasive deployment and usage. Conclusion In this paper, I have examined the importance of the location of AI companies and, by so doing, introduced the politics of locationality as a concept to extend our understanding and explain the standing that nations have in AI development and how this translates to the influence they wield in AI governance. My discussion showed that although Africa has a growing AI startup ecosystem, a burgeoning user market base and state influence in policymaking, the continent still faces a mix of disadvantages in historical, technological, political, legal, economic, labour and data spheres. This mix reveals the locational disadvantage that confronts Africa as a place from which AI ethics, cultures and governance can be shaped in the global sense. It relates to Hassan’s (2023) point on the ‘lack of African AI innovations that are rooted in the local context but [have] the potential to compete at the global scale’ (p. 1430). What I found was that AI companies that have global scale and presence are headquartered mainly in leading tech countries (the US, China and European nations) and have subjected themselves to the governing laws of these countries. Conversely, the laws and policies of leading tech countries mandate that these companies, among other things, embed the respective country’s ethos, values and cultures into AI systems. The link between AI systems and cultural values as outlined in this paper, therefore, points to the connection between media and cultural studies and AI. In particular, the paper highlights the importance of considering how AI is produced and shaped by cultural understandings, and how AI further projects, shapes and extends culture. It suggests the need for greater research emphasis on the connections between media, AI and culture, given that AI represents the intersection of human labour, machine learning, infrastructure (e.g. data centres), mineral resources (e.g. lithium), media inputs (vast troves of data: information, texts, pictures, videos, maps, and codes) and classification systems that all combine to transform AI models into mechanics of knowledge construction (Crawford, 2021). The outcome are mediated outputs (e.g. AI responses to prompts and automated decisions) – outputs that simultaneously represent and transform culture, even as they are shaped by it. Equally crucial and more important, in the context of this paper, is the need to consider the locationality of it all. By this I mean the globalised and complex interaction of locations in AI development and governance that potentially determines whose culture is elevated and whose culture is progressively silenced in a relationship defined by hegemony and inequality. Key to understanding this hegemony-inequality mix is the principle of locational advantage, which Iammarino and McCann (2013) describe as a two-way relationship between multinationals and their locational headquarters, noting that just as location is becoming increasingly vital for multinationals, so also are multinationals progressively more important for location (i.e. cities or countries). It underscores my argument on the politics of locationality, which shows that just as large AI firms need locations, such as the US or China, that have the right mix of advantages to serve as headquarters, so also do leading tech countries need and use these firms in setting the cultural and governance paradigm for AI technologies. This structural makeup reflects the imbalances in AI development, explaining why AI governance systems exist within institutional-heterogenous regimes that reproduce Global North dominance (Png, 2022). Addressing these imbalances is what African countries, supported by the AU, should focus on, and there are initial signs that African countries are starting to do this (as seen in the AI strategies). However, a bolder and more robust approach to AI development that is attuned to Africa’s unique sociopolitical context and economic evolution is needed to confront the imbalances tied to the politics of locationality. By so doing, Africa can be positioned to assume locational advantage, with important consequences for inclusive AI and the contribution of global majority countries to the cultures and values embedded in AI systems and the ethics and frameworks by which they are governed. AcknowledgmentsThe author extends appreciation to the editors and reviewers for their helpful and constructive comments, and also to the participants at the 2023 International Symposium on AI Cultures at the University of Turin for their feedback – all of which strengthened the article.Declaration of conflicting interestsThe author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.FundingThe author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Leverhulme Trust as part of an Early Career Fellowship. The University of Sheffield provided open access funding.ORCID iDVincent Obia https://orcid.org/0000-0003-1650-9103ReferencesAccess Partnership (2024) Tech Policy Trends 2024. Available at: https://accesspartnership.com/tech-policy-trends-2024-the-global-south-is-the-new-epicentre-of-internet-governance-innovation/ (accessed 4 January 2025).African Union (2024) Continental artificial intelligence strategy. Available at: https://au.int/sites/default/files/documents/44004-doc-EN-_Continental_AI_Strategy_July_2024.pdf (accessed 10 October 2024).AI Directory (2024) Database of artificial intelligence & machine learning companies. Available at: https://www.aidirectory.org/ (accessed 10 September 2024).Bergére C (2019) “Don’t tax my megabytes”: Digital infrastructure and the regulation of citizenship in Africa. International Journal of Communication 13: 4309–4326.Bradford A (2023) Digital Empires: The Global Battle to Regulate Technology. New York, NY: Oxford University Press. Crossref.Chan A, Okolo CT, Terner Z, et al. (2021) The limits of global inclusion in AI development. arXiv. Crossref. PubMed.CIPESA (2022) Which way for data localisation in Africa? Available at: https://cipesa.org/wp-content/files/briefs/Which_Way_for_Data_Localisation_in_Africa___Brief.pdf (accessed 16 December 2024).Companies Market Capitalisation (2024) Largest AI companies by market capitalization. Available at: https://companiesmarketcap.com/artificial-intelligence/largest-ai-companies-by-marketcap/#google_vignette (accessed 5 September 2024).Crawford K (2021) The Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence. New Haven: Yale University Press.Cyberspace Administration of China (2022) Provisions on the management of deep synthesis of internet information services. Available at: https://www.cac.gov.cn/2022-12/11/c_1672221949354811.htm (accessed 12 October 2024).Darktrace (2024) Master services agreement. Available at: https://cdn.prod.website-files.com/626ff4d25aca2edf4325ff97/66fc2bc0317776db72ac418c_Darktrace%20MSA%20v2.0%202024-10-01.pdf (accessed 4 October 2024).DCByte (2023) Africa’s key data centre markets. Available at: https://africadca.org/wp-content/uploads/2023/07/Title_Africas-Key-Data-Centre-Markets.pdf (accessed 17 December 2024).Dunning JH (1998) Location and the multinational enterprise: A neglected factor? Journal of International Business Studies 29(1): 45–66. Crossref. Web of Science.Egypt AI Strategy (2021) Egypt National Artificial Intelligence strategy. Available at: https://mcit.gov.eg/Upcont/Documents/Publications_672021000_Egypt-National-AI-Strategy-English.pdf (accessed 5 July 2023).Eke DO, Wakunuma K, Akintoye S (2023) Introducing responsible AI in Africa. In: Eke DO, Wakunuma K, Akintoye S (eds) Responsible AI in Africa: Challlenges and opportunities. Chan, Switzerland: Palgrave Macmillan, pp.1–11. Crossref.Emerging Technology Observatory (2024) AI Country Activity Tracker. Available at: https://cat.eto.tech/?countryGroups=Africa%2CAsia%20Pacific%2CEurope%2CLatin%20America%20and%20the%20Caribbean%2CNorthern%20America%2COceania&dataset=Patent&expanded=Summary-metrics (accessed 7 September 2024).Epoch AI (2024) Notable AI models. Available at: https://epoch.ai/data/notable-ai-models?view=table#explore-the-data (accessed 7 September 2024).European Union (2024) Artificial Intelligence Act. Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L_202401689 (accessed 25 January 2025).Fratini S, Musiani F (2024) Data localization as contested and narrated security in the age of digital sovereignty: The case of Switzerland. Information, Communication & Society: 1–19. Crossref. Web of Science.Gemini (2024) User Agreement. Available at: https://www.gemini.com/legal/user-agreement#section-business-day (accessed 4 October 2024).Giovane CD, Ferencz J, López-González J (2023) The nature, evolution and potential implications of data localisation measures. OECD Trade Policy Papers, No. 278. Paris, France: OECD Publishing.Global Partnership on Artificial Intelligence (2025) GPAI Members. Available at: https://gpai.ai/community/ (accessed 7 January 2025).Gondwe G (2023) CHATGPT and the Global South: How are journalists in sub-Saharan Africa engaging with generative AI? Online Media Global Communication 2(2): 228–249. Crossref.Hassan Y (2023) Governing algorithms from the South: A case study of AI development in Africa. AI & Society 38: 1429–1442. Crossref. Web of Science.Henshall W (2023) How China’s new AI rules could affect U.S. companies. Times, 19 September.Iammarino S, McCann P (2013) Multinationals and Economic Geography: Location, Technology and Innovation. Cheltenham: Edward Elgar. Crossref.Jungco KG (2024) 100 top AI companies trendsetting in 2024. Available at: https://www.datamation.com/featured/ai-companies/ (accessed 9 September 2024).Lee H-K (2022) Rethinking creativity: Creative industries, AI and everyday creativity. Media, Culture & Society 44(3): 601–612. Crossref. Web of Science.Lin B (2024) Beyond authoritarianism and liberal democracy: Understanding China’s artificial intelligence impact in Africa. Information, Communication & Society 27(6): 1126–1141. Crossref. Web of Science.Ludec CL, Cornet M, Casilli AA (2023) The problem with annotation. Human labour and outsourcing between France and Madagascar. Big Data & Society 10(2): 1–13. Crossref.Maslej N, Fattorini L, Perrault R, et al. (2024) The AI index 2024 annual report. Available at: https://aiindex.stanford.edu/wp-content/uploads/2024/05/HAI_AI-Index-Report-2024.pdf (accessed 23 September 2024).Mauritius AI Strategy (2018) Maritius Artificial Intelligence strategy. Available at: https://ncb.govmu.org/ncb/strategicplans/MauritiusAIStrategy2018.pdf (accessed 20 June 2023).Mhlambi S, Tiribelli S (2023) Decolonizing AI Ethics: Relational Autonomy as a Means to Counter AI Harms. Topoi 42: 867–880. Crossref. Web of Science.Moorosi N (2024) Better data sets won’t solve the problem — we need AI for Africa to be developed in Africa. Nature 636: 276. Crossref. PubMed. Web of Science.Muldoon J, Wu BA (2023) Artificial intelligence in the colonial matrix of power. Philosophy & Technology 36(80): 1–24.Natale S, Guzman AL (2022) Reclaiming the human in machine cultures: Introduction. Media, Culture & Society 44(4): 627–637. Crossref. Web of Science.National AI Commission Act (2023) National AI Commission Act (H.R.4223). Available at: https://www.congress.gov/bill/118th-congress/house-bill/4223/text (accessed 4 October 2024).Nayebare M (2019) Artificial intelligence policies in Africa over the next five years. XRDS 26(2): 50–54 Crossref.Nigerian AI Strategy (2024) National Artificial Intelligence strategy [draft]. Available at: https://ncair.nitda.gov.ng/wp-content/uploads/2024/08/National-AI-Strategy_01082024-copy.pdf (accessed 4 October 2024).Nuwer R (2024) Africa’s newest resource could be a game-changer for the global south. Nature, 19 September.Okolo CT, Aruleba K, Obaido G (2023) Responsible AI in Africa—Challenges and opportunities. In: Eke DO, Wakunuma K, Akintoye S (eds) Responsible AI in Africa: Challenges and Opportunities. Chan, Switzerland: Palgrave, pp.35–64. Crossref.OpenAI (2024) Terms of Use. Available at: https://openai.com/policies/row-terms-of-use/ (accessed 4 October 2024).Ott BL, Mack RL (2014) Critical Media Studies: An Introduction. West Sussex: Wiley Blackwell.Parks L, Thompson R (2020) The slow shutdown: Information and Internet regulation in Tanzania from 2010 to 2018 and impacts on online content creators. International Journal of Communication 14: 4288–4438. Web of Science.Plantinga P, Shilongo K, Mudongo O, et al. (2024) Responsible artificial intelligence in Africa: Towards policy learning. Data & Policy 6: e72. Crossref. Web of Science.Png M-T (2022) At the tensions of South and North: Critical roles of Global South stakeholders in AI governance. In: Bullock JB, Chen Y-C, Himmelreich J, et al. (eds) The Oxford Handbook of AI Governance. Oxford: Oxford University Press. Crossref.Qwen (2023) Alibaba cloud international website terms of use. Available at: https://www.alibabacloud.com/help/en/legal/latest/alibaba-cloud-international-website-terms-of-use-alibaba-cloud-international-website-terms-of-use (accessed 8 October 2024).Rwandan National AI Policy (2022) The national AI policy. Available at: https://rura.rw/fileadmin/Documents/ICT/Laws/Rwanda_national_Artificial_intelligence_Policy.pdf (accessed 5 July 2023).Sargsyan T (2016) Data localization and the role of infrastructure for surveillance, privacy, and security. International Journal of Communication 10: 2221–2237. Web of Science.Tidjon LN, Khomh F (2022) The different faces of AI ethics across the world: A principle-implementation gap analysis. arXiv. Crossref.UBTECH (2023) Terms of Use. Available at: https://www.ubtrobot.com/en/privacy/termOfUse (accessed 4 October 2024).US Executive Order (2023) Executive order on the safe, secure, and trustworthy development and use of artificial intelligence. Available at: https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-the-safe-secure-and-trustworthy-development-and-use-of-artificial-intelligence/ (accessed 10 January 2024).Wong PN (2021) Techno-Geopolitics: US-China Tech War and the Practice of Digital Statecraft. London: Routledge. Crossref.World Economic Forum (2025) AI Governance Alliance - Our partners. Available at: https://initiatives.weforum.org/ai-governance-alliance/partners#countries=Africa (accessed 7 January 2025).

In the Middle East, China seeks to appear neutral by engaging in dialogue with all actors, from Tehran to Tel Aviv, including Riyadh. Beijing's objective is clear: to establish itself as a guarantor of stability to secure its energy and commercial interests. The People's Republic of China's (PRC) commitment to the Middle East reflects a carefully calibrated geostrategic approach, aiming to preserve regional stability, ensure uninterrupted access to energy resources, and promote its famous Belt and Road Initiative (BRI), also known as the 'New Silk Roads.' In this region, the Sino-Iranian relationship is most often highlighted due to its political weight and military dimension; however, whether in economic, diplomatic, or strategic terms, Beijing's presence in this area certainly extends beyond its ties with Tehran. As elsewhere in the world, in the Middle East, China divides its diplomatic partnerships into several types, listed here in descending order of intensity: "global strategic partnerships" (in the Middle East: Egypt, Iran, Saudi Arabia, United Arab Emirates, and Bahrain); "strategic partnerships" (Iraq, Jordan, Kuwait, Oman, Qatar, Syria, Turkey, and the Palestinian Authority); "friendly cooperation partnerships" (Lebanon and Yemen); and finally, "innovative global partnerships" (Israel). China is redefining its priorities in the Middle East  Energy dependence is one of the essential drivers of China's policy in the region. Gulf countries, particularly Saudi Arabia, the United Arab Emirates, and Kuwait, are among the main suppliers of crude oil to the PRC—well ahead of Iran in terms of volume, as well as reliability and investment opportunities. This economic reality compels China to invest much more heavily in the Gulf Cooperation Council (GCC) countries, where financial stability, political predictability, and institutional openness promote sustainable strategic partnerships and infrastructure development, unlike in Iran. For these reasons, the Gulf countries have indirectly become essential nodes in the architecture of the BRI through trade, receiving nearly six times more Chinese investments than Iran. Although Iran has geographical interest as a potential corridor between China and Europe, the persistent effect of international sanctions, poor economic management, and regional adventurism severely limit its ability to attract sustainable Chinese investments. In contrast, Israel offers a functioning and favorable environment for investors, making it a preferred destination for Chinese capital and infrastructure projects. Its economy is based on solid infrastructure and a dynamic technology sector. Despite geopolitical shocks, including the recent clashes with Iran – which will undoubtedly impact the country's business environment – Israel, as a member of the OECD, remains attractive in terms of business environment and investment, mainly because the foundations of its economic environment are stronger than those of Iran or Saudi Arabia. The Hebrew state occupies an increasing place within the BRI – not due to the volume of Chinese direct investments, but because of its strategic position and ambitious infrastructure agenda. Its geographical location – straddling Europe, Asia, and Africa via the Mediterranean – makes it a key land and maritime connectivity point that provides an alternative option to Iran for linking China to Europe. In this context, China and Israel have made progress in their negotiations for a free trade agreement that has been discussed since 2016. Although discussions have been suspended since 2023, there remains hope that an agreement could be signed once stability returns to the region. China's trade with Israel and Iran: divergent trajectories China's commercial relations with Israel and Iran reveal two distinct economic approaches.  With Israel, trade is increasingly marked by imports of high technologies - particularly in the semiconductor field - reflecting a growing technological interdependence.  In contrast, trade with Iran remains focused on industrial goods and natural resources, illustrating a more traditional partnership based on access to raw materials. These contrasting dynamics reflect China's strategic flexibility, which relies on Israeli innovation while securing its supplies from Iran.  Tensions in the Strait of Hormuz are testing China's strategy in Iran  For nearly fifty years, Iran has turned to China for economic support. However, despite 21 Chinese greenfield investment projects between 2003 and 2020 – mainly in the energy sector – Beijing has gradually begun to disengage due to international sanctions and ongoing regional instability. Large companies such as CNPC and Sinopec have reduced their stake, or even abandoned some projects, while tech companies like Huawei and Lenovo have also scaled back their presence in the Iranian market. Between 2017 and 2019, Iran reportedly saw an outflow of Chinese capital estimated at $990 million. China views instability in the Middle East, particularly the confrontation between Israel and Iran, as a direct threat to its economic and commercial interests. As the world's largest buyer of Iranian oil, China is especially concerned about potential disruptions in strategic maritime routes, particularly the Strait of Hormuz. In June 2025, Chinese oil imports from Iran significantly increased, reaching up to 1.8 million barrels per day – a surge that occurred just before the military escalation between Israel and Iran, which raised concerns about the security of the Strait of Hormuz. A sign of the weight of the PRC in this region: after the reciprocal airstrikes between Israel and Iran, and Tehran's threat to close this crucial passage, U.S. Secretary of State Marco Rubio urged Beijing to intervene, emphasizing China's heavy dependence on this oil route. China–United States: strategic tensions in a transitioning Middle East China and the United States recently signed a trade agreement and established a 90-day tariff truce, aiming to resolve some key disputes and stabilize economic relations. Furthermore, on the military front, Iran and China signed a military cooperation agreement as early as 2016, reflecting their mutual desire to counter American influence in the region and secure trade routes. Historically, China has supported Iran through arms sales, technology transfers, and training programs. Moreover, despite the official rhetoric, some reports suggest that Chinese technologies have contributed to the development of Iran's ballistic program. The Sino-Iranian relationship in defense illustrates a two-level diplomacy: China discreetly supports Iran's military autonomy while ensuring it does not compromise its strategic relationships with other important regional partners, including Israel, with whom it also maintains top-level security dialogues. This highlights Beijing's broader ambition: to avoid regional polarizations and maintain a balance of power favorable to its interests. China's strategy in the Middle East towards the United States is based on a principle of equidistance: to increase its influence without direct confrontation, while drawing strategic benefits from Saudi Arabia, Israel, Iran, and Egypt. In this perspective, the People's Republic of China does not seek to replace the United States as the dominant power in the region, but rather to offer a multipolar presence focused on preserving stability. Maintaining the status quo, avoiding direct conflicts, and building a parallel system of influence through infrastructure, trade, and diplomacy are the pillars of China's position in the Middle East. This approach ensures Beijing secure access to energy, trade corridors, and sustainable geopolitical influence from Tel Aviv to Tehran.

Abstract European states currently lack a clear joint strategy on China and a coordinated approach to US–China competition. This article offers a novel perspective on the challenges for European approaches to this issue due to an omnipresent transatlantic component and the risk of an alliance dilemma. Illustratively focusing on France, Germany and the UK, it demonstrates that Europeans are facing a transatlantic alliance dilemma with the risks of abandonment and entrapment. It argues that Europe needs to strike a balance between its dependence on Washington, especially with regard to European security, while fearing entrapment by the US approach towards Beijing as it aims to maintain economic ties with China. The article concludes that the ramifications of this dilemma can be mitigated through a distinctly European approach to China, strengthening European coordination on China and bolstering European strategic autonomy. As a conceptual piece rather than a full empirical analysis, this article therefore unpacks the strategic challenge and lays the groundwork for further empirical works on the topic. Introduction Strategic competition between the US and China plays out in many realms of international affairs, ranging from global trade to security in the Indo-Pacific. European states are directly affected by this dynamic as they maintain critical ties with both sides. Albeit allies of the US through NATO, Europeans have been reluctant to align with the US on its approach to the Indo-Pacific and China, which is currently characterised by the quest to win the strategic competition with Beijing in all areas of international affairs (see Leoni 2023). Furthermore, Europe maintains close economic ties with Beijing, and imports from China to the EU have most recently increased (Lovely and Yan 2024). European governments certainly do not pursue an approach of maintaining equidistance between the US and China: not only do they regularly emphasise their strategic proximity to Washington, but more recent events, such as the willingness of European allies to publicly adopt the wording of the communiqué from NATO’s Washington summit (NATO 2024) describing China as an ‘enabler of Russia’s war effort in Ukraine’, clearly demonstrate that the tone is changing in European capitals (Politi 2023). However, Europeans still tend to emphasise China’s role in global affairs and the importance of including it in multilateral cooperation formats. With the re-election of Donald Trump as US president, European policy on China and its approach to US–China competition will increasingly be a focus of the US administration. While the exact approach of the new US government still remains to be defined, there are good reasons to anticipate a more hawkish China policy from Washington, through which the US might seek not only to further compete with China, but to actually win this competition (see Pottinger and Gallagher 2023). When designing their approaches towards China—which, among the key European states, only Germany has done so far, with the publication of its China strategy in 2023—Europeans always face a transatlantic elephant in the room. US–China competition, a structuring feature of international relations shaping the global order today through the increasing emergence of geopolitical blocs (see Leoni and Tzinieris 2024), and China’s rising global influence in almost all areas of international affairs—ranging from climate to economics, the rules-based order and security—are forcing European states to reflect on their approach vis-à-vis Beijing (for a full discussion, see Oertel 2023; García-Herrero and Vasselier 2024). As NATO members, European states also need to adapt their strategy in light of the partnership with the US as their key ally. This article argues that European approaches towards China, as shown in the examples of France, Germany and the UK, have a distinctly transatlantic component. It illustrates how these three European states find themselves in an alliance dilemma with the US, and how the risks associated with alliances also define European approaches to China and US–China competition more broadly. As the US administration regularly refers to China as a ‘challenge’ (US Department of Defense 2022), this article alludes to this formulation through the coining of the term ‘transatlantic China challenge’ to describe the strategic challenges Europeans are facing with regard to defining their approach vis-à-vis China and US–China competition more broadly. It offers a conceptual understanding of the strategic challenges for Europe in this context and thereby constitutes a basis for a more thorough empirical analysis. The alliance dilemma and European strategy in US–China competition Originating in realist international relations theory, the alliance dilemma generally describes a situation in which states face risks resulting from joining an alliance. As demonstrated by Snyder (1984), smaller allies especially face a parallel risk of abandonment and entrapment by a hegemon, that is, the dominating power, after joining an alliance. Abandonment, in these circumstances, implies that the hegemon has no further interest in defending or supporting the smaller allies, whereas entrapment refers to a situation in which a state is ‘dragged into a conflict over an ally’s interests that [it] does not share, or shares only partially’ (see Snyder 1984, 466–8). In the context of alliances, a small state is ‘the weaker part in an asymmetric relationship, which is unable to change the nature or functioning of the relationship on its own’ (Wivel et al. 2014, 9), and hence has more limited space for action than the great powers (Wivel and Thorhallsson 2018, 267). This definition arguably applies to Europe in its partnership with the US, as demonstrated by the excessive military and economic dominance of the US as compared to the European states (see Stockholm International Peace Research Institute n.d.; International Monetary Fund 2025). The re-election of Trump as US president now presents the risk of an increased alliance dilemma for Europeans. On the one hand, Trump has announced several times that he does not value the alliance commitments within NATO and potentially would not defend European allies (Sullivan 2024), threatening Europe with abandonment. This scenario is being taken seriously in European capitals, and reflections on how ‘defending Europe with less America’ (Grand 2024) could shape up have gained traction, especially in 2024. Similarly, defence initiatives within the EU to enhance the European contribution to the continent’s security have leapt forward in recent years (see Scazzieri 2025). On the other hand, even the Biden administration had pushed Europe to align with the US approach on China (see Lynch et al. 2023). However, France and Germany in particular, as the big EU member states, have been hesitant to do so, as reflected in France’s opposition to the opening of a NATO liaison office in Tokyo (McCurry 2023) and Germany’s vote against tariffs on Chinese electric vehicles, fearing reprisals from Beijing (Demarais 2024). Trump’s foreign policy might be strongly characterised by issue linkage, which means that policies in one area will be linked to those in another area. Through this strategy, the new US administration might force Europeans into alignment and thereby entrap them, making them accept policies they are not eager to support (see Barkin and Kratz 2025). The exact policies of the Trump administration vis-à-vis European allies remain to be seen, but it is not hard to imagine a scenario in which abandonment and entrapment could emerge or increase, namely when the threat of abandonment is used to entrap allies and force them to support certain policy decisions. The alliance dilemma could play out for Europeans specifically when designing their approaches towards China (see Barkin and Kratz 2025) and formulating their response to US–China competition more generally. As noted above, among the big European states, only Germany has formally adopted a strategy on China, in 2023 (The Federal Government of Germany 2023). However, China and the response to US–China competition takes a prominent place in France’s Indo-Pacific Strategy and its strategy review (Government of France 2021; Secrétariat général de la défense et de la sécurité nationale 2022), and the UK systematically included the dimension of strategic competition in its Integrated Review and its refresh (Government of the United Kingdom 2021, 2023) and has announced an ‘audit’ of its China policy under the new government (Taylor 2024). While these strategies emphasise their individual approaches towards China and the risks stemming from US–China competition, the US has increasingly pressured Europe to align with its approach (Lynch et al. 2023) and can be expected to continue this pressure (Barkin and Kratz 2025). Through the potential issue linkage of security (openly questioned by President Trump) and China policy, Europe therefore finds itself in a new form of alliance security dilemma. European approaches to US–China competition: strategic hedging How can Europe respond to the alliance dilemma of the risks of abandonment and entrapment when it comes to its approaches to China? Reviewing the theoretical literature on the alliance dilemma, one can imagine different strategies. According to Snyder, members of alliances can choose between strategies that strengthen or weaken their commitment to the alliance. To demonstrate commitment, actions could include reassurances for the ally or demonstrations of loyalty, whereas actions to weaken the commitment to the alliance could consist of restraining the ally (mostly to reduce the risk of entrapment in a conflict), increasing bargaining power over the ally or preserving options for realignment outside the existing alliance (for a full discussion, see Snyder 1984, 466–9). Alternative strategies include hiding from cooperation, that is, ‘seeking to maximize autonomy by opting out of specific aspects of the cooperation or by setting up “bastions” in the cooperation’, or bandwagoning, through which states pursue strategies of adaptation ‘to the more powerful actors in the cooperation’ (Pedersen 2023, 442). At the moment, it seems that France, Germany and the UK ‘drive on sight’ rather than approaching the question holistically. The following analysis aims to unpack how the three European states see US–China competition, the risk of the alliance dilemma and how these reflections have played out so far in their strategies. The strategies of France, Germany and the UK on China demonstrate that their approaches are influenced by a distinctly transatlantic component and reflect the transatlantic alliance dilemma. This is visible in their (1) high awareness of the risks stemming from great power competition, (2) approaches to managing the risk of short-term abandonment, and (3) hedging to mitigate the medium- and long-term risks of abandonment and entrapment. The empirical evidence for this analysis was gathered through a qualitative analysis of European strategic documents, statements and policy decisions taken mostly during the period of the Biden administration. However, in light of the risk of a scaling-up of the alliance dilemma under the Trump administration, sources and evidence accessible by the end of January 2025 were included to illustrate the European approaches. In addition to publicly available documents and the sources mentioned above, this paper draws on conversations with policymakers and experts under the Chatham House rule. Mitigating risks from US–China competition: multilateralism instead of alignment That France, Germany and the UK are close allies with the US is clearly visible in their respective strategies on China, not least because of references they make to the importance of the alliance and their descriptions of their own positions between the two great powers. Overall, France, Germany and the UK share the perception of US–China competition and the emergence of blocs as potentially harmful to their interests. As a consequence, all three call for an inclusive multilateral order instead of falling into a logic of blocs, as the increasing competition is seen as a risk for Europe (Secrétariat général de la défense et de la sécurité nationale 2022, 9–15; The Federal Government of Germany 2020, 24–6; Government of the United Kingdom 2023, 22–6). The response of all three European powers to the emergence of blocs is multilateralism: instead of clearly aligning with the US, the French, German and British strategies call for building broader multilateral coalitions, which should, eventually, also include China (The Federal Government of Germany 2020, 23–6; Government of the United Kingdom 2023). The tone in Paris, Berlin and London towards Beijing has clearly changed over recent years; accordingly, the European capitals were also willing to support strong wording on China in the 2024 NATO summit declaration, which describes China as a ‘critical enabler’ of Russia’s ongoing war against Ukraine (NATO 2024). Albeit also recognised as a critical partner for key issues such as climate policy and trade, European states openly describe China as a ‘systemic rival’ and occasionally call out China’s behaviour, as they did, for example, in the case of a note verbale on the South China Sea (UN 2020). Nevertheless, Europe has not (yet) given in to US pressure to align with Washington’s more confrontational approach towards China (Etienne 2024). Even if European states and Washington have moved closer to each other, especially on economic security (Meyers and Reinsch 2023), the European positions on US–China competition demonstrate that Europeans are not willing to fully endorse or follow Washington’s approach—not least because European imports from China have increased in recent years (Lovely and Yan 2024). Managing the risk of short-term abandonment Since Trump’s election, the risk of abandonment by the US has been seen as increasingly high in Paris, Berlin and London.1 This is not least because Trump has openly questioned his willingness to adhere to Article 5 of the North Atlantic Treaty in the case of an armed attack on Europeans (Sullivan 2024). Europeans are especially concerned about issue linkage in this context, meaning that demands in a policy area other than security could be used as a condition. Concretely, Trump could use the threat of abandonment as leverage to compel Europe to align with the US on China policy.2 Barkin and Kratz (2025) suggest that Europe adopt a ‘carrot and stick’ approach, whereby Europe could start with an offer to the US: buying more liquified natural gas, defence goods and agricultural products from the US could mitigate the risk of abandonment. However, there is awareness among European states that coercion from the US to align on US–China policy, especially when linked to the threat of abandonment, might best be mitigated through enhancing European military capabilities—which would still leave the continent exposed to these threats, though to a lesser extent.3 Addressing the risk of medium- and long-term abandonment and entrapment: transatlantic hedging However, the risk of at least partial abandonment is not a new challenge for European strategy, and had already infused earlier strategic thinking. The shift of US strategic priorities away from Europe and to the Indo-Pacific has already been demonstrated in the allocation of resources to the different theatres. Moreover, European states have increasingly become aware that US forces will be withdrawn from their territories in the future and have concluded that they will have to step up their own commitment to European security (see Grand 2024). In parallel, there is an awareness in European capitals that showing more interest in the Indo-Pacific and giving more importance to policy on China is also a way for Europeans to demonstrate an understanding of their ally’s strategic priorities.4 Accordingly, the approaches of France, Germany and the UK to China and the Indo-Pacific also have to be understood as a commitment to the priorities of the US in order to keep this ally engaged in the European theatre and mitigate the risk of abandonment; however, European states abstain from fully aligning with the US approach, as their capabilities and strategic goals are perceived as diverging from those of the US. In this way, Europe aims to avoid entrapment over the medium term through slightly distancing itself from US policy. While all three European states also call for de-risking from China and diversification of their supply chains, maintaining strong economic ties with Beijing is a key component of their respective approaches—which contrasts with the US calls for decoupling. Furthermore, these states have never formally endorsed the US policy on China (Etienne 2024). Nevertheless, enhancing European capabilities would not only send a signal to Washington, but also qualify as hedging, understood as an ‘insurance policy’ to avoid a deterioration in US–Europe relations if the US opted for abandonment, or even as part of a move towards emancipation to reduce strategic dependencies on Washington (see Fiott 2018, 4–6). Conclusion: a transatlantic China challenge Designing their approaches to China and to US–China competition more broadly constitutes a complex strategic dilemma for European states. Paris, Berlin and London do not fully align with Washington’s approach, and it remains to be seen whether they will be willing to do so under the second Trump administration. To manage the risk of abandonment and entrapment, European states pursue different individual approaches to strategic hedging: their strategies on China and US–China competition are designed in a way that allows them to mitigate the risk of abandonment which might stem from significant transatlantic divergence, and to avoid automatic entrapment through their slight distancing from the US approach. From a theoretical perspective, this article has demonstrated that the alliance dilemma, along with the parallel fear of abandonment and entrapment by the US, is a major factor accounting for Europe’s limited strategies on China and its hedging behaviour. This article offers a conceptual analysis of the structural forces explaining European strategies, but other strategic cultures and relationships with the US could offer important complementary insights. To further analyse how individual European states design their strategies vis-à-vis China in light of the alliance dilemma and potential domestic constraints and specificities, neoclassical realism could offer an interesting analytical concept. This approach posits that structural forces set the parameters for foreign policy and treats domestic factors, including strategic culture, as intervening variables (see Rose 1998). Accordingly, it appears well suited for foreign policy analysis, and has indeed gained popularity in the field in recent years (see, for example, Martill and Sus 2024; Meibauer et al. 2021; Weber 2024). Empirically, this article constitutes a conceptual starting point rather than an exhaustive analysis of the strategy-making processes of European states with regard to China and US–China competition and makes a more comprehensive assessment desirable. The findings of this article have broader implications for policymaking. First, they demonstrate the necessity for Europe to determine its place in the increasing US–China competition. European coordination on the respective approaches vis-à-vis the US—especially in light of potential coercion to align—and China is of paramount importance to ensure that foreign policy strategies are mutually reinforcing and not undermining European objectives. Second, the article demonstrates that Europe currently responds to the ‘transatlantic China challenge’ through transatlantic hedging: while this strategy seems to be promising in the short term, it is questionable to what extent the strategy is sustainable and could help European states to navigate the parallel challenges of abandonment and entrapment. Unless Europe decides to fully align with the US—and it is questionable whether this decision would be in its interest—European states would be well advised to develop a sustainable long-term approach to China. A transatlantic dialogue on China, in which Europe and the US openly discuss synergies and divergences, could help prevent misunderstandings and decrease the risk of coercion or issue linkage due to a misreading of European approaches in Washington. Third, as the risks of (at least partial) abandonment and entrapment are systemic challenges due to the current composition of the transatlantic alliance, a logical step for European states to decrease their dependence on the US as the hegemon in the alliance would be to significantly strengthen European capabilities. Stronger military capabilities could help mitigate the ramifications of abandonment, and the aforementioned distinctly European strategy could allow Europe to avoid strategic entrapment in relation to China imposed by Washington. As Europe remains the junior partner in the transatlantic alliance, the parallel risks of abandonment and entrapment, as well as issue linkage, are highly likely to influence its approaches towards China in the long term, but there are certainly ways to render this ‘transatlantic China challenge’ less challenging. ORCID iDGesine Weber https://orcid.org/0009-0008-2643-0400Footnotes1. Conversation with French, German and British experts in Berlin, January 2025.2. Conversation with French, German and British experts in Berlin, January 2025.3. Conversation with French, German and British experts in Berlin, January 2025; conversation with European experts and officials in Paris, January 2025.4. Conversation with officials from Germany and France in Paris, November 2024; conversation with French, German and British experts in Berlin, January 2025.ReferencesBarkin N., Kratz A. (2025). Trump and the Europe–US–China Triangle. Rhodium Group, 16 January. https://rhg.com/research/trump-and-the-europe-us-china-triangle/. Accessed 18 January 2025.Demarais A. (2024). Divided we stand: The EU votes on Chinese electric vehicle tariffs. European Council on Foreign Relations, 9 October. https://ecfr.eu/article/divided-we-stand-the-eu-votes-on-chinese-electric-vehicle-tariffs/. Accessed 25 January 2025.Etienne P. (2024). The European Union between the United States and China: Should we choose between equidistance and following? Fondation Robert Schuman, 8 October. https://www.robert-schuman.eu/en/european-issues/763-the-european-union-between-the-united-states-and-china-should-we-choose-between-equidistance-and-following. Accessed 22 December 2024.Fiott D. (2018). Strategic autonomy and the defence of Europe. European Union Institute for Security Studies, Brief 12/2018. https://www.iss.europa.eu/sites/default/files/EUISSFiles/Brief%2012__Strategic%20Autonomy.pdf. Accessed 25 January 2025.García-Herrero A., Vasselier A. (2024). Updating EU strategy on China: Co-existence while de-risking through partnerships. Mercator Institute for China Studies. https://merics.org/en/external-publication/updating-eu-strategy-china-co-existence-while-de-risking-through-partnerships. Accessed 30 January 2025.Government of France. (2021). France’s Indo-Pacific strategy. https://www.diplomatie.gouv.fr/IMG/pdf/en_dcp_a4_indopacifique_022022_v1-4_web_cle878143.pdf. Accessed 31 January 2025.Government of the United Kingdom. (2021). Global Britain in a competitive age: The integrated review of security, defence, development and foreign policy. London: The Stationery Office. https://www.gov.uk/government/publications/global-britain-in-a-competitive-age-the-integrated-review-of-security-defence-development-and-foreign-policy. Accessed 18 January 2025.Government of the United Kingdom. (2023). Integrated review refresh 2023: Responding to a more contested and volatile world. London: The Stationery Office. https://www.gov.uk/government/publications/integrated-review-refresh-2023-responding-to-a-more-contested-and-volatile-world. Accessed 18 January 2025.Grand C. (2024). Defending Europe with less America. European Council on Foreign Relations, 3 July. https://ecfr.eu/publication/defending-europe-with-less-america/. Accessed 18 January 2025.International Monetary Fund. (2025). World economic outlook: GDP data mapper [Map]. https://www.imf.org/external/datamapper/NGDPD@WEO/OEMDC/ADVEC/WEOWORLD. Accessed 23 January 2025.Leoni Z. (2023). Grand strategy and the rise of China: Made in America. Agenda Publishing.Leoni Z., Tzinieris S. (2024). The return of geopolitical blocs. Survival, 66(2), 37–54.Lovely M. E., Yan J. (2024). As the US has relied less on imports from China, the EU has imported more. Pederson Institute for International Economics, 24 August. https://www.piie.com/research/piie-charts/2024/us-has-relied-less-imports-china-eu-has-imported-more. Accessed 22 December 2024.Lynch S., Toosi N., Moens B., Banco E. (2023). The U.S. wants Europe to stand up to China. Europe says: Not so fast. Politico, 3 August. https://www.politico.com/news/2023/03/08/us-europe-china-00086204. Accessed 18 January 2025.Martill B., Sus M. (2024). Winds of change? Neoclassical realism, foreign policy change, and European responses to the Russia-Ukraine War. British Journal of Politics & International Relations.McCurry J. (2023). France opposed to opening of Nato liaison office in Japan, official says. The Guardian, 7 June. https://www.theguardian.com/world/2023/jun/07/france-opposed-to-opening-of-nato-liaison-office-in-japan-official-says. Accessed 25 January 2025.Meibauer G., Desmaele L., Onea T., Kitchen N., Foulon M., Reichwein A., Sterling-Folker J. (2021). Forum: Rethinking neoclassical realism at theory’s end. International Studies Review, 23(1), 268–95.Meyers E., Reinsch W. A. (2023). The push for U.S.–EU convergence on economic security policy. Center for Strategic and International Studies, 7 July. https://www.csis.org/analysis/push-us-eu-convergence-economic-security-policy. Accessed 26 January 2025.NATO. (2024). Washington Summit declaration. https://www.nato.int/cps/cn/natohq/official_texts_227678.htm. Accessed 18 January 2025.Oertel J. (2023). Ende der China-Illusion: Wie wir mit Pekings Machtanspruch umgehen müssen. Munich: Piper Verlag.Pedersen R. B. (2023). Small states shelter diplomacy: Balancing costs of entrapment and abandonment in the alliance dilemma. Cooperation and Conflict, 58(4), 441–59.Politi A. (2023). The paradigm shift in EU–China relations and the limits of the EU’s current strategy towards China: A relational perspective. Asian Affairs 54(4), 670–93.Pottinger M., Gallagher M. (2024). No substitute for victory: America’s competition with China must be strategic and ideological. Foreign Affairs, 10 April. https://www.foreignaffairs.com/united-states/no-substitute-victory-pottinger-gallagher. Accessed 18 January 2025.Rose G. (1998). Neoclassical realism and theories of foreign policy. World Politics, 51(1), 144–72.Scazzieri L. (2025). Towards an EU ‘defence union’? Centre for European Reform, 30 January. https://www.cer.eu/publications/archive/policy-brief/2025/towards-eu-defence-union. Accessed 30 January 2025.Secrétariat général de la défense et de la sécurité nationale. (2022). Revue nationale stratégique 2022. https://www.sgdsn.gouv.fr/publications/revue-nationale-strategique-2022. Accessed 18 January 2025.Snyder G. H. (1984). The security dilemma in alliance politics. World Politics, 36(4), 461–95.Stockholm International Peace Research Institute. (n.d.). SIPRI Military Expenditure Database.Sullivan K. (2024). Trump says he would encourage Russia to ‘do whatever the hell they want’ to any NATO country that doesn’t pay enough. CNN, 11 February. https://edition.cnn.com/2024/02/10/politics/trump-russia-nato/index.html. Accessed 29 January 2025.Taylor R. (2024). UK government policy towards China. https://lordslibrary.parliament.uk/uk-government-policy-towards-china/. Accessed 18 January 2025.The Federal Government of Germany. (2020). Policy guidelines for the Indo-Pacific. www.auswaertiges-amt.de/blob/2380514/f9784f7e3b3fa1bd7c5446d274a4169e/200901-indo-pazifik-leitlinien–1–data.pdf. Accessed 21 December 2024.The Federal Government of Germany. (2023). China-Strategie der Bundesregierung. Berlin: Auswärtiges Amt. https://www.auswaertiges-amt.de/blueprint/servlet/resource/blob/2608578/810fdade376b1467f20bdb697b2acd58/china-strategie-data.pdf. Accessed 18 January 2025.UN (2020). Note verbale, UK NV No. 162/20, 16 September. https://www.un.org/Depts/los/clcs_new/submissions_files/mys_12_12_2019/2020_09_16_GBR_NV_UN_001.pdf. Accessed 25 January 2025.US Department of Defense. (2022). National Defense Strategy of the United States of America. https://media.defense.gov/2022/Oct/27/2003103845/-1/-1/1/2022-NATIONAL-DEFENSE-STRATEGY-NPR-MDR.pdf. Accessed 25 January 2025.Weber G. (2024). Zeitenwende à la française: Continuity and change in French foreign policy after Russia’s invasion of Ukraine. British Journal of Politics & International Relations.Wivel A., Bailes A. J. K., Archer C. (2014). Setting the scene: Small states and international security. In Archer C., Bailes A. J. K., Wivel A. (eds.), Small states and international security: Europe and beyond (pp. 3–25). London: Routledge.Wivel A., Thorhallsson B. (2018). Brexit and small states in Europe: Hedging, hiding or seeking shelter? In Rosamond B., Nedergaard P., Diamond P. (eds.), The Routledge handbook of the politics of Brexit, 1st edn. (pp. 266–77). Abingdon-on-Thames: Routledge.

In recent years, US leadership has embraced techno-nationalism amid geopolitical and technological rivalry with China, aiming to minimise reliance on imported chips from Asia. These components are crucial for producing consumer goods, military hardware, and AI systems. The United States set the ambitious goal of developing a self-sufficient semiconductor supply chain during Donald Trump’s first term, and continued under Joe Biden. There is consensus in the United States on the critical role of unfettered access to chips when it comes to ensuring economic and national security. It is unlikely that this technological policy dynamic will undergo significant shifts in the foreseeable future. Despite a shared objective among both Republicans and Democrats to revive the US semiconductor industry, their approaches diverge significantly. Donald Trump has his own vision for advancing this sector, one that contrasts sharply with Joe Biden’s strategy. For instance, Trump has criticised aspects of Biden-era initiatives, including the 2022 CHIPS and Science Act, which he has called counterproductive. Trump, on the other hand, favours a more aggressive tariff policy and a reduction in federal spending, arguing that major tech companies can do well without additional government support. The future balance of power—both technological and geopolitical—among the key global actors will be shaped by the development trajectory of the US semiconductor industry. Biden’s semiconductor legacy The United States holds a dominant position in chip design, but maintains a relatively modest share in global semiconductor manufacturing—just 10 percent according to SIA data in 2022, and slightly up to 11 percent according to 2025 data provided by TrendForce research firm. Major US tech giants like Nvidia or Qualcomm remain heavily reliant on chips produced in Taiwan. This dependency has increasingly been seen as unacceptable by US leadership, especially in the context of the ongoing tech war with China. Washington now views such reliance as a significant national security risk. During Donald Trump’s first presidential term, the decision was made to attract leading chip manufacturers—most notably Taiwan Semiconductor Manufacturing Company (TSMC), the world’s largest contract chipmaker—to set up operations in the United States. This initiative proved successful: in 2020, TSMC agreed to invest $12 billion to build a chip fabrication plant in Arizona (Fab 21).   The Biden administration continued Trump’s push to revitalise the semiconductor industry. In August 2022, the CHIPS and Science Act was passed, allocating about $53 billion in government subsidies for the semiconductor sector, along with tax incentives to encourage both foreign and domestic firms to establish chip manufacturing operations on US soil. Additionally, the CHIPS for America programme was introduced to address several key goals, namely, to secure a stable supply chain for both cutting-edge and legacy semiconductors, to reinforce US leadership in R&D, and to boost employment, as investment in the chip industry was expected to generate hundreds of thousands of new jobs in microelectronics-related fields. Biden’s programme has borne fruit. Major chipmakers have launched large-scale construction of fabs across the United States. In 2022, Intel started building a $28 billion facility in Ohio; Samsung initiated two plants in Texas worth about $40 billion; and TSMC decided to expand its Arizona site to three modules, increasing its total investment from $12 billion to $65 billion. According to TSMC CEO C.C. Wei, the Arizona facility began mass production in the fourth quarter of 2024 using its N4 (4nm class) process technology, with performance comparable to its fabs in Taiwan. This marks the most advanced semiconductor production facility currently operating in the United States. Plans are in place to launch a second module for 3nm chip production by 2028, followed by a third module by 2030, which will manufacture 2nm and 1.6nm chips and their variants. The Biden team aimed for the United States to capture 20 percent of global advanced chip manufacturing by 2030. Democrats have adopted a comprehensive approach to rebuilding the semiconductor industry not just focusing on building advanced fabs, but also investing in support areas such as chip testing and packaging, materials production, and R&D. A substantial $13 billion in federal funds has been earmarked for these purposes. For instance, grants and loans were used to support GlobalFoundries’ plans to build an advanced packaging and photonics centre in New York State. Arizona State University also received significant support from the US Department of Commerce, including a $100 million allocation for research and development in next-generation chip packaging technologies. Wide geographic distribution is a striking feature of the emerging US semiconductor supply chain (Figure 1). Key activities are being established across numerous states: Oregon (semiconductor manufacturing), Idaho (semiconductor and material manufacturing), Utah (semiconductor manufacturing), Montana (equipment manufacturing), Colorado (semiconductor and material manufacturing), New Mexico (packaging), Kansas (semiconductor manufacturing and packaging), Louisiana (equipment manufacturing), Missouri (materials), Minnesota (semiconductor manufacturing),Michigan (materials),Indiana (packaging and semiconductor manufacturing), Ohio (materials and semiconductor manufacturing), Vermont (semiconductor R&D and manufacturing), Pennsylvania (materials), North Carolina (semiconductor manufacturing), Georgia (materials and semiconductor manufacturing), and Florida (materials and semiconductor manufacturing). Among these, several states stand out for their significance and comprehensive involvement: California (semiconductor manufacturing and R&D), Arizona (semiconductor, equipment, and material manufacturing, packaging, R&D), Texas (semiconductor and material manufacturing, packaging, R&D), and New York (materials, semiconductor manufacturing, and R&D).   According to a 2024 study by the Boston Consulting Group commissioned by the Semiconductor Industry Association (SIA), over 90 projects have been launched in 28 states since the CHIPS Act was passed, totalling nearly $450 billion in private investment. However, the Biden administration did not pursue full semiconductor self-sufficiency as a goal. There was recognition that recreating the entire supply chain domestically would, even at the initial stage, require a vast amount of time and financial resourcesНадпись: MichiganНадпись: IndianaНадпись: Pennsylvania estimated at around $1 trillion. Therefore, US policymakers have advocated for a collective semiconductor supply chain among allies and partners by building international alliances. In 2022, the Unite States proposed creating the CHIP 4 alliance (United States, South Korea, Japan, and Taiwan), which, with coordinated efforts, could have become a dominant force in the semiconductor industry capable of influencing nearly every segment of the global value chain, with the exception of assembly and testing, where mainland China currently plays a leading role. In this way, Trump’s initiative to revive the semiconductor industry has not only continued under Biden, but evolved into a more ambitious and costly programme. The SIA, in its above report, painted an optimistic picture for the future of the US semiconductor sector. It projects that chip manufacturing capacity in the United States will triple over the next decade (2022–2032), growing by 203 percent. This expansion is expected to require $646 billion in investment, or 28 percent of global capital spending in the semiconductor industry. As a result, the United States could increase its share of global chip production from the current 10 percent to 14 percent by 2032. Additionally, experts estimate that the new projects will create over 58,000 new jobs in the semiconductor sector and hundreds of thousands more in related industries.   Despite its ambitious nature, the initial phase of Biden’s semiconductor programme has revealed several challenges. The industry has run into numerous internal obstacles slowing the construction of manufacturing facilities, including a shortage of skilled labour, high labour and construction material costs, bureaucratic hurdles (e.g., obtaining environmental permits), slow disbursement of promised subsidies by the US authorities, union-related delays, cultural differences, and more. These issues have caused delays in launching chip fabrication plants, thereby slowing the pace at which the US can achieve relative technological autonomy in the rapidly evolving semiconductor field. For example, TSMC postponed the start of mass production at the first module of Fab 21 from 2024 to 2025, and delayed the second module from 2026 to 2027–2028. Intel’s costly attempt to reclaim leadership in advanced chip manufacturing has strained its budget, forcing the company to delay its Ohio fab launch from 2025 to 2030. Samsung, initially planning to start production in Texas in the second half of 2024, pushed the timeline to 2025. These delays in fab construction also impacted the schedules of launching supplier plants, including chemical and material producers like LCY Chemical, Solvay, Chang Chun Group, KPPC Advanced Chemicals (Kanto-PPC), and Topco Scientific. The external component of the Biden administration’s technology policy has also failed to develop as envisioned. After several years of existence, the CHIP 4 has failed to become a multilateral coordination mechanism, and its potential members have not assumed any binding commitments. Only one virtual meeting was held in 2023. The reason lies in internal disagreements within the alliance and concerns about various risks, including geopolitical ones. Under the Biden administration, the United States made a strong start in the semiconductor sector, launching a wide range of fab construction projects and attracting billions of dollars in public and private investment. However, the process of reviving the US semiconductor industry has proven slower than anticipated. Government subsidies have been disbursed sluggishly, with some companies yet to receive their funding, and the construction of many high-tech industrial facilities has been postponed. Moreover, Biden overestimated the willingness of US allies and partners to join formal technological alliances. Trump’s radical approach To encourage both domestic and foreign chip suppliers to set up manufacturing in the United States, Donald Trump, in contrast to Joe Biden, chose coercion (tariffs) over incentives (government subsidies). Criticising his predecessor’s CHIPS Act, Trump argued that companies didn’t need money, but rather motivation in the form of import tariffs ranging from 25 percent to 100 percent. In his view, such measures would compel businesses to invest in US chip manufacturing, especially since these companies have the financial capacity and, therefore, don’t need to rely on government funding. Almost immediately after taking office, Trump threatened chip manufacturers with higher tariffs. At first glance, this move might seem economically illogical. Why, for instance, punish TSMC—a key partner of major US fabless companies like Nvidia, Apple, and Qualcomm—especially when there is no comparable alternative, either in the United States or globally? Even Intel, despite its struggles, depends on wafers from the Taiwanese firm (its import dependency is about 30 percent). Yet despite apparent lack of logic, the “stick” approach proved effective. In early March 2025, TSMC announced plans to invest approximately $100 billion to build three new fabs for high-performance semiconductor wafers, two advanced chip packaging plants, and one R&D centre. This raises the question: did the world’s largest chipmaker really get spooked by Trump’s tariff threats and, therefore, decide to make an unprecedented investment in the US economy? In theory, TSMC—sitting in the centre of the global microelectronics industry—could have passed tariff-related costs on to its American clients, who would have had little choice but to continue purchasing its products due to the lack of viable alternatives. Furthermore, a significant share of TSMC’s semiconductors is not shipped directly to the United States, but instead follows a supply chain tour through Asia, where the bulk of chip packaging, testing, and electronics assembly occurs (this infrastructure is only just beginning to take shape in the United States, and that process is anything but fast). Analysts at Bernstein suggest that political pressure, rather than tariffs themselves, drove TSMC’s decision. That assessment holds some merit, but it appears that a combination of factors was at play. First, TSMC itself is interested in expanding its global presence. Taiwan’s Minister of Economic Affairs Kuo Jyh-Huei commented on TSMC’s $100 billion investment in the US semiconductor sector by saying, “TSMC already has plants in the United States and Japan, and is now building a new one in Germany. This has nothing to do with tariffs. TSMC’s global expansion is a major development.” Similarly, in 2020 during Trump’s first term, company representatives said that the decision to build a plant in Arizona was “based on business needs.” Indeed, the move offers several benefits to TSMC, including increased company capitalisation and minimised risks in the event of conflict with mainland China or natural disasters (earthquakes are not uncommon in Taiwan). Second, the United States remains TSMC’s primary market, and the tariff threat did play its part. In Taiwan, there’s an understanding that when Trump talks about higher tariffs, he isn’t bluffing, because his seriousness was evident during his first term and was experienced first-hand by Canada and Mexico. On April 2, 2025, nearly the entire rest of the world—including Taiwan—faced a new wave of tariffs, with Taiwanese exports to the United States hit by a 32 percent duty (though semiconductors were not yet affected). A 100-percent tariff on semiconductors is unlikely, as it would significantly damage the market value of US tech firms. Still, protective barriers on semiconductors are expected—Trump’s administration has promised to implement them in the coming months. These measures aim to level the production cost of chips between the United States and Taiwan, thereby enhancing the competitiveness of US-made semiconductors. And finally, TSMC, together with the Taiwanese authorities, is not willing to mar relations with the United States for political reasons. This became evident from TSMC’s earlier decision to support US sanctions against mainland China by refusing to supply its most advanced chips manufactured using 7nm and more sophisticated process technologies even though that market had been a significant source of profit. After TSMC announced plans to expand its presence in the United States, the Trump administration decided to take more radical action and to scrap the CHIPS and Science Act, a signature achievement of the Biden administration. However, some Republican members of Congress are calling for the law to be preserved, albeit with certain amendments. Trump’s hands are not completely untied in this regard, so it is unlikely he can ignore Congress’s position. Even if the legislation gets amended, the process will likely be drawn out, as the CHIPS and Science Act received bipartisan support and has many supporters among Republicans. Another strategically important issue for the Trump administration is the competitiveness of domestic manufacturers. According to the Taiwanese leadership, TSMC will continue to expand operations in Taiwan, and the most advanced semiconductor technologies will not leave the country. For “the most powerful AI chips in the world to be made right here in America” efforts will be needed on the part of national champions—and soon. In 2025, the leader in producing the most advanced 2nm chips will be determined. The main contenders in this race are TSMC, Samsung, and Intel. Intel, however, finds itself in a difficult position. The company has been facing serious financial troubles for several years and lags behind competitors in mastering cutting-edge production processes. The year 2024 was one of Intel’s most challenging: it underwent a major restructuring (creating a separate chip manufacturing unit, Intel Foundry), posted record losses of $18 billion, and saw a significant drop in its stock price. As a result, about 15 percent of the workforce, including CEO Pat Gelsinger, was laid off; dividend payments were suspended; and a sweeping cost-cutting plan was launched, including deep cuts in capital expenditures over the coming years and a scaling back of global expansion plans. According to Intel Products CEO Michelle Johnston Holthaus, the company failed to capitalise effectively on the artificial intelligence boom and continues to fall behind its competitors technologically. Although Intel plans to begin 18A (2nm) chip production in 2025, there are no guarantees of competitiveness in power efficiency, performance, yield rate, cost, or timely mass production. In March, media reported that Nvidia and Broadcom began testing certain chip components, but such testing, of course, does not guarantee Intel will secure orders. Apparently, the Trump administration itself has doubts about the US company’s capabilities, as it has proposed that TSMC acquire shares in Intel Foundry. Negotiations with the Asian manufacturer began only in February 2025, meaning they are still at a very early stage.   What short-term challenges does the Trump administration face in revitalising the US semiconductor industry? Technological lag There is a high likelihood that the United States will continue to lag behind Taiwan for several years in the production of advanced semiconductors. TSMC plans to begin producing chips using a 1.4nm process by 2028, while on US soil—if deadlines aren’t pushed back again—the Taiwanese firm will only be producing 3nm chips by that time. Although some hope is being placed on Intel, there is no guarantee that the American champion will be able to compete with TSMC, or that a potential collaboration with TSMC (if it acquires a stake in Intel Foundry) will be successful. Inadequate production capacity Experts estimate that the output capacity of TSMC’s factories under construction in Arizona is less than one-fifth of the company’s 5nm and 3nm capacity in Taiwan. According to analysts at Bernstein Research, with the deployment of additional production in Arizona, the United States could raise its self-sufficiency in advanced chip production to 40-50 percent between 2030 and 2032. In the near term, this would only cover about half of the chip demand from US tech giants. Moreover, TSMC has not specified clear timelines or technologies for its US expansion. Intel could partly close the gap, but that depends on how competitive its chips are and how quickly it can overcome its financial difficulties. Slow rollout of production facilities TrendForce forecasts that the US share of global advanced chip production could grow from 11 percent to 22 percent by 2030. However, the construction of TSMC’s first Arizona plant took nearly five years, and there are no guarantees that future factories will be built fast enough to double US chip output by 2030. Labour shortage Developing a relatively self-sufficient microelectronics ecosystem requires a highly skilled workforce. However, the United States is facing severe staff shortages. By 2030, estimates suggest a shortfall of 67,000 to 90,000 professionals in the semiconductor field. China’s response to US sanctions The United States is not the only country leveraging interdependence in the semiconductor industry as a tool of pressure. China is responding in kind, though currently in a relatively restrained manner. In 2024, the Chinese government decided to completely ban exports of gallium, germanium, antimony, and ultra-hard materials to the United States even though the restrictions apply only to direct shipments. These actions not only drive up raw material prices (e.g., antimony prices more than tripled since early 2024), but also force US authorities to consider domestic mining and search for alternative suppliers abroad. High production costs According to the SIA, building and operating chip fabs in the United States is 30 to 50 percent more expensive than in Asia. Unofficial reports suggest that chips made at Fab 21 in Arizona cost 10 percent to 30 percent more than their Taiwanese counterparts (more precise figures are not publicly available). The high cost is attributed to expensive construction of facilities, high salaries (US engineers earn three times more than their Taiwanese counterparts, incomplete domestic semiconductor supply chains (some materials must still be imported)—TSMC CEO has complained about it—and complex logistics (finished wafers often need to be sent back to Taiwan or elsewhere for packaging).70 Even if tariffs eventually equalise chip pricing, US fabless companies like Apple or Nvidia may still find it more economical to source chips from Asia, where a properly functioning semiconductor ecosystem already exists—unlike in the United States, where such infrastructure is still in its infancy. Trump’s current tariff policy Imposing tariffs could lead to a significant increase in prices for components, equipment, and materials, while also injecting uncertainty into the semiconductor industry. For instance, it remains unclear how semiconductor manufacturers will operate under new tariffs on imported chip-making equipment sourced from the EU, Japan, South Korea, and Taiwan. The cost of such equipment can reach hundreds of millions of dollars—for example, the latest Low-NA EUV lithography machine from Dutch company ASML is priced at $235 million. If Intel, TSMC, and other firms are required to pay import duties of 20 percent or more, chip manufacturing in the United States will become prohibitively expensive, undermining investment plans of the manufacturers that have committed to building advanced fabs on American soil. Naturally, US officials understand that sharp moves in semiconductor policy—such as an aggressive tariff regime—carry significant risk and could spark a true technological crisis. In April 2025, the US Department of Commerce’s Bureau of Industry and Security (BIS) launched an investigation under Section 232 of the Trade Expansion Act of 1962 to determine the impact of semiconductor imports and related equipment on national security. Interested parties submitted comments, many urging extreme caution in this highly sensitive sector, which depends on a complex global supply chain split across multiple phases and countries. Thus, SIA recommended that any tariffs be phased in gradually to allow the US industry to continue functioning efficiently until domestic production capabilities are fully established. The US Chamber of Commerce called for restraint, warning that comprehensive tariffs on the semiconductor supply chain could damage US industry and undermine cooperation with allies and partners in achieving key national security goals. The Chamber also noted that foreign semiconductor companies have made long-term investment commitments to build capacity in the United States, and that political uncertainty and instability could jeopardise the stated goal of re-shoring semiconductor supply chains. *** As TSMC founder Morris Chang once said, America’s effort to ramp up its own chip production may well prove to be “a very expensive exercise in futility.” Microelectronics is one of the most complex industries in the world requiring not only massive financial investment, but also time. For decades, the industry developed within the framework of global division of labour. Now, building a relatively self-sufficient supply chain within a single country could take just as long. Yet, in the medium and long term, America’s push to revive its semiconductor industry may prove justified. The United States holds a strong position in the sector, and US companies control about 50 percent of the global semiconductor market. Furthermore, the United States remains a powerful magnet for talent, and possesses vast financial and political resources. Some experts believe that over time, the United States could weaken Taiwan’s dominance as the global hub of advanced chip manufacturing. The resurgence of the US semiconductor industry will reshape the global technological order in three key ways. First, it will trigger a transformation of the global semiconductor supply chain. Second, it will lead to greater US independence from imports of critical technologies, which means erosion of importance of some players in the industry, weakening their “technological shield”. Finally, it will cement US technological superiority in many critical industries, from AI to military systems, accelerating a global technological divide with profound geopolitical consequences. Indeed, America has the potential to become one of the world’s leading semiconductor production centres, provided that several key conditions are met, such as a favourable geopolitical environment, domestic political stability, and the absence of disruptive black swan events. However, Trump’s risky tariff policy could trigger unpredictable cascading effects, both domestically (e.g., higher prices for electronics and microelectronics products) and internationally (e.g., retaliatory tariffs by US trade partners), posing serious threats for the US semiconductor industry. First published in the Valdai Discussion Club.