By Jasurbek Khamrakulov, Undergraduate student at UWED, intern at IAIS   Background In recent years, the relations between Europe and China have entered an increasingly complex and controversial phase. While the European Union is determined to engage with China in open dialogue and cooperate on common challenges, including the climate challenge, it has raised serious concerns over trade imbalances, supply chain vulnerabilities and technological competition. The EU identifies China simultaneously as a “partner, competitor, and systemic rival,” an expression first introduced in its 2019 strategic outlook and reaffirmed by EU member states in 2023. China remains one of the EU’s largest trading partners, with bilateral trade in goods reaching €732 billion in 2024. However, the EU has run trade deficit with China for decades. It amounted €359.9 billion in 2025 and surpassed the €312.2 billion deficit of 2024. China became the EU's fourth-largest partner for exports and remains its biggest for imports. Chinese companies exported the goods to Europe with the value of €559.5 billion. Besides, EU imports of manufactured goods made up 97.3% of total imports from China. The most imported manufactured goods were machinery(54.4%), other manufactured goods (33%) and chemicals (9.8%). De-risking strategy Therefore, the central theme of current EU policy has become “de-risking” strategy. The term was introduced in the strategy of European Commission President Ursula von der Leyen during her first mandate. She emphasized the need to mitigate of dependencies on China, particularly in areas of high tech and dual-use goods.  Even under the pressure of the American worldwide tariffs imposed by the Trump’s administration in 2025, von der Leyen viewed Europe and China as two of the world's largest markets, designed to support reformed free trading system, thereby rejecting Trump’s demands for decoupling from China.  Eventually, the situation in the rare earth market intensified after China expanded export controlson several strategic minerals, such as lithium, cobalt, and rare earth minerals, which are essential for electric vehicles, renewable energy technologies, semiconductors, and defense industries. This exposed Europe’s vulnerability in global supply chains. Brussels responded by introducing “economic security doctrine” and initiating new measures focused on stockpiling, recycling, joint purchasing, and diversifying imports through partnerships with alternative suppliers. An important step was taken with the EU and the United States have signing a new strategic partnership on critical minerals aimed to reduce Western dependence on China. The agreement, signed by EU Trade Commissioner Maroš Šefčovič and U.S. Secretary of State Marco Rubio, includes coordination on mining, refining, recycling, stockpiling, and trade policies related to critical raw materials. In this context, the de-risking strategy is clearly seen in European Union’s accelerated efforts to reduce its dependence on China for critical raw materials. These initiatives build upon the EU’s Critical Raw Materials Act, which aims to ensure that no more than 65% of a strategic raw material comes from a single external source.  Moreover, the core elements of the new doctrine included improving coordination among member states and accelerating the use of anti-dumping and anti-subsidy measures. Initiatives such as “RESourceEU” were prioritized to strengthen domestic industrial capacity in sectors like batteries, AI, defense, and semiconductors. However,  one of the major obstacles facing Europe’s approach comes down to the lack of funding, domestic mining capacity and processing infrastructure to compete with Chinese companies. Maintaining the approach would require greater use of financing from the European Investment Bank for funding and to the EU's Global Gateway plan, its own version of China's Belt and Road scheme. Despite the hurdles, the EU has maintained additional tariffs on electric vehicles manufactured in China. The tariffs, first introduced in 2024 after an anti-subsidy investigation, can reach up to 35.3% depending on the manufacturer. However, the EU has also demonstrated a pragmatism by allowing individual companies to negotiate tariff exemptions through minimum-price agreements and import quotas. In February 2026, the European Commission approved the first such exemption for Volkswagen’s China-made Cupra Tavascan model, showing that Europe seeks to manage economic competition with China rather than choose the path of immediate decoupling. At the same time, the European Commission recommended the member states exclude of equipment from Huawei and ZTE as part of a broader strategy to strengthen cybersecurity and reduce strategic vulnerabilities linked to “high-risk suppliers” in key sector, such as critical digital infrastructure. The move demonstrates how the European rhetoric is going far beyond trade into the digital security, where technological dependence is now viewed as a geopolitical risk. This process of securitization is also evident in the EU’s climate policies and protectionism. A major example is the Carbon Border Adjustment Mechanism (CBAM), which entered into force in January and applies carbon pricing to imports of emissions-intensive goods such as steel, aluminium, and cement. Since China is a major exporter of these products, the mechanism creates additional compliance costs and competitive pressure for Chinese firms, while encouraging cleaner production standards. Geopolitical dimensions One of the main sensitive obstacles to ease EU-China relations lies in the geopolitical tensions.  China’s close relationship with Russia remains a major concern for European policymakers, especially regarding sanctions circumvention. Meanwhile, China’s expanding influence in the Global South, especially in regions rich in critical minerals such as Latin America and Africa, also overlaps with Europe’s own diversification strategies. As a result, European companies are gradually adjusting their business models through supplier diversification, localization strategies, and closer geopolitical risk assessment. The engagement of the EU with China has long been shaped by the trends and development of Transatlantic relations. Over the last several months, the growing tensions between Europe and the United States have created opportunities for China to re-engage with Europe, but on terms favorable to Beijing. Chinese analysts increasingly interpret Europe’s push for “strategic autonomy” as a sign that the EU may distance itself from Washington and soften its strategy towards China. However, Europe’s de-risking policies are not simply the result of American pressure, but come from the EU’s own assessment of strategic vulnerabilities linked to Chinese supply chains, state subsidies and technological dependencies. As the pressure from the US rapidly grows, China will likely attempt to exploit transatlantic tensions by offering selective economic incentives and pressure Europe to loosen its economic restrictions. European and Chinese perceptions Chinese experts often describe Europe’s policy as an over-securitization of economic relations that damages the EU’s own economic interests. They argue that reducing dependence on Chinese supply chains increases production costs and complicates the continent’s green and digital transitions by limiting access to affordable Chinese technologies and industrial goods. Some media channels even claim that Europe’s de-risking is turning into a form of “de-development”caused by pressure from the United States rather than by Europe’s own economic interests. From the constructivist perspective, European perceptions of China are becoming extremely complex and contradictory. While many Europeans support reducing economic dependence on China and favor diversifying trade relations, Chinese exports to Europe continue to grow rapidly, especially in major economies such as Germany, France, and Italy. Public opinion reflects a tension between concern over European deindustrialization and consumers’ attraction to affordable Chinese products. These divisions are also geographical, with Southern European countries generally more open to Chinese economic engagement than more industrialized northern states. Many European governments continue pursuing pragmatic engagement with China despite Brussels’ tougher regulatory approach. Ongoing visits by European leaders, including Friedrich Merz’s visit in February are viewed as evidence that many EU member states still value economic cooperation and recognize the importance of Chinese markets. A harder line on China is gaining momentum inside the European Commission. It is actively exploring the unused Anti-Coercion Instrument (ACI) that would include tariffs, import/export restrictions, bans on public procurement participation, restrictions on intellectual property rights, to counter Chinese economic pressure. A key debate among the 27 Commissioners on the EU’s China strategy is scheduled for 29 May 2026. This shift follows frustration with Beijing’s lack of meaningful concessions despite repeated EU concerns and Chinese retaliation threats over EU legislation such as the “Made in Europe” Industrial Accelerator Act. Despite the Commission’s tougher stance, member states remain divided. Germany’s Chancellor F. Merz has expressed the intention of a long-term trade deal with Beijing. Spain continues to pursue close economic ties, while France and Belgium call for a firmer approach to protect European industries. Activating stronger tools like the ACI would require qualified majority support. Policy recommendations Given the potential risks and internal inconsistencies, a key takeaway for the EU should be continuing a balanced and pragmatic de-risking strategy rather than full economic decoupling from China. Firstly, the EU must accelerate the diversification of strategic supply chains. Priority sectors include critical raw materials, semiconductors, batteries, pharmaceuticals, and green technologies. This requires rapidly expanding trade and investment agreements with alternative suppliers in Latin America (particularly MERCOSUR), Africa, and the Indo-Pacific to reduce exposure to geopolitical shocks, export controls, and economic coercion. Secondly, Europe needs to strengthen its domestic industrial and technological base through targeted industrial policy. The Critical Raw Materials Act, Carbon Border Adjustment Mechanism (CBAM), and green industrial subsidies should be complemented by increased funding for research and development, faster permitting for strategic projects, and infrastructure modernization. While pursuing these goals, the EU must avoid excessive protectionism that could raise consumer costs, slow down innovation, or trigger damaging reactions. Instead, it should rely on precise, WTO-compatible tools such as anti-subsidy measures, investment screening, and cybersecurity standards in high-risk sectors. Thirdly, the EU should overcome internal fragmentation by developing a more unified and coherent China policy. National differences have too often undermined collective decision-making. Achieving consensus among key member states will significantly enhance the Union’s negotiating position. At the same time, the EU should deepen strategic partnerships with Indo-Pacific democracies, including Japan, South Korea, India, and Australia, in critical minerals, technology standards, digital governance, and maritime security. These partnerships will help reduce over-reliance on China without undermining the free and open trading system. Ultimately, Europe’s approach toward China is best understood not as a move toward confrontation, but as an attempt to find the proper way in the anarchic international system through pragmatic balancing. The potential success can be only achieved if Europe’s de-risking strategy represents both a response to geopolitical pressures and a long-term effort to reach strategic autonomy in the multipolar order. And that strategic autonomy begins with unity; without it, Europe negotiates with China not as a power, but as a collection of voices. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.

By Khabibulla Khayrullaev, UWED undergraduate, intern at IAIS   Origins and Institutional Framework On January 20, 2025, the first day of his second term, President Donald Trump signed Executive Order 14158, establishing the Department of Government Efficiency (DOGE). Structured as a temporary division embedded within the existing United States Digital Service (renamed the United States DOGE Service), the organization was assigned an 18-month mandate with a formal termination date of July 4, 2026 – the 250th anniversary of American independence, which Trump called “the perfect gift to America”. Entrepreneur Elon Musk, who had donated more than $290 million to Trump’s 2024 campaign — the largest individual political donation in American history — led DOGE as a “special government employee”,  a status permitting a maximum of 130 working days per year. DOGE’s stated mission was to modernize federal information technology, maximize productivity, and eliminate waste, fraud, and abuse. In practice, teams of young technology professionals, many recruited directly from Musk’s own companies or straight out of college, were deployed across dozens of federal agencies with broad authority to cancel contracts, terminate grants, and facilitate mass layoffs, operating with limited interagency coordination and frequently without prior legal review.   The Savings Claim: What DOGE Says vs. What Analysts Find DOGE’s official “wall of receipts”, a public tally on its website, claims between $160 and $215 billion in savings as of early 2026. Musk’s original public promise of $2 trillion was subsequently revised downward to $1 trillion, then further to $150–200 billion — a reduction of more than 90% from the initial target. A February 2026 CBS News investigation found that the 13 largest claimed cancellations were all incorrect, and journalists identified billions of dollars in miscounting. A rigorous counter-analysis by the nonpartisan Partnership for Public Service estimated that DOGE’s actions — including paying tens of thousands of employees for months of unworked leave under the ‘Fork in the Road’ deferred-resignation program, litigation costs, and precipitous productivity losses — will cost taxpayers $135 billion in fiscal year 2026 alone. The Yale Budget Lab separately calculated that the reduction of IRS staffing by roughly 40% will cost the federal government $323 billion in lost tax revenue over the next decade due to reduced auditing capacity and lower compliance rates. In a deposition made public in March 2026, DOGE employee Nate Cavanaugh acknowledged that cost-cutting efforts fell far short of the original $2 trillion goal. The deposition was part of a lawsuit filed by the American Council of Learned Societies, alleging that DOGE used OpenAI’s ChatGPT to algorithmically identify and cancel more than $100 million in diversity, equity, and inclusion grants, a practice critics described as using automation to circumvent congressional authorization. In total, DOGE claimed to have canceled 13,440 contracts and eliminated the roles of more than 300,000 federal employees.   The Human Toll: Federal Workers Left Behind Between January 2025 and January 2026, 386,826 workers departed the federal government, including approximately 17,000 through formal reductions in force and many thousands more through DOGE’s ‘Fork in the Road’ deferred-resignation offer. Simultaneously, only 122,000 new workers joined the federal workforce – a 55% decrease from 2024 hiring levels. Among those fired, 24,000 were subsequently rehired following court orders finding the terminations unlawful, including bird flu researchers at the Department of Agriculture who had been dismissed without apparent awareness of their role. An NBC News investigation published April 22, 2026 spoke with 13 former federal employees and found that seven remain fully unemployed. The organization WellFed, which supports displaced federal workers, estimates that only 25% of its members have found new employment. Among former USAID workers, the group OneAID estimates at least 50% remain unemployed. Unemployment benefits have lapsed for many; healthcare coverage has become precarious. One former IRS employee described the 2026 tax filing season as likely the most difficult "since the pandemic," citing severe understaffing and mounting backlogs.   The GAO Report and Data Security Failures (April 28, 2026) In what represents one of the most significant institutional accountability findings of the DOGE era, the Government Accountability Office (GAO) published a report on April 28, 2026 concluding that the Treasury Department granted a DOGE associate access to sensitive federal payment systems, systems that process trillions of dollars in government transactions, without fully following its own security controls. The GAO confirmed that DOGE personnel did not always comply with Treasury’s established protocols either. The DOGE employee central to the GAO‘s findings was Marko Elez, who resigned from Treasury on February 6, 2025after the public disclosure of racist social media posts and subsequently went on to work for DOGE at other agencies. The GAO found that Treasury’s data loss prevention tools failed to block Elez from improperly sending unencrypted information on foreign aid to DOGE associates at the General Services Administration. Rep. Richard Neal, the senior Democrat on the Ways and Means Committee, stated that “GAO has confirmed our worst fears”.  The GAO confirmed it is working on additional audits of DOGE access across other government systems.   Musk’s Departure and What Remained Elon Musk officially departed his role on May 29, 2025, upon the expiration of his 130-day “special government employee” term. His departure was accompanied by a post on X stating: “The @DOGE mission will only strengthen over time as it becomes a way of life throughout the government”.  White House press secretary Karoline Leavitt confirmed that DOGE employees embedded in agencies as political appointees would remain: “The DOGE leaders are each and every member of the president’s cabinet, and the president himself”. Approximately 100 DOGE employees remain embedded across federal departments, having transitioned from their temporary DOGE roles into permanent political appointments. The administration has simultaneously launched a Gen Z federal hiring initiative – the “US Tech Force” — to rebuild federal technology capacity using young software engineers and AI specialists recruited from Silicon Valley. The Office of Personnel Management has contracted with Workday, Inc.to process federal retirement applications, marking a significant step toward the privatization of core governmental functions. At the same time, the administration’s proposed 2027 federal budget seeks to nearly halve NASA’s budget from $7.3 billion to $3.9 billion, while DOGE has claimed to have worked with the National Science Foundation to cancel DEI-related grants. A 57% majority of Americans, according to a Washington Post–ABC–Ipsos poll, disapprove of the way Musk handled his role in the Trump administration. About six in ten say they are worried the president has done too much to cut the size of the federal government.   The July 4 Terminus: End of an Organization, Continuation of a Project DOGE is scheduled to formally dissolve on July 4, 2026, coinciding with Trump’s proposed “Great American Fair” celebrating the 250th anniversary of the Declaration of Independence. The formal termination of the organization, however, will not end the restructuring it has initiated. The civil service hiring criteria introduced under DOGE, which require applicants to submit essays articulating alignment with the administration’s policy goals, represent a fundamental departure from the merit-based system established by the Pendleton Civil Service Reform Act of 1883. The AI Deregulation Decision Tool, designed to analyze and eliminate half of the federal government’s 200,000+ regulations by January 2027, continues its work. The DOGE experiment has demonstrated that the American federal bureaucracy can be disrupted at speed and at scale through a combination of executive authority and private-sector aggressiveness. It has also demonstrated the institutional costs of disruption without adequate process: billions of dollars in litigation, hundreds of thousands of displaced workers, degraded service capacity in agencies from the IRS to the VA, and now a formal GAO finding of security failures in the handling of the government’s most sensitive financial systems. Whether the restructuring that emerges from this process produces a more efficient state or merely a more politically compliant one – remains the defining question of DOGE’s contested legacy. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.

  By Ismoilova Soliha, UWED undergraduate, intern at IAIS   The parliamentary elections in Slovenia, held on March 22, concluded with a narrow margin between the leading political forces. The «Freedom Movement» (Gibanje Svoboda, GS), headed by Prime Minister Robert Golob, secured approximately 28.5-28.66% of the vote and obtained 29 seats in parliament. While, the Slovenian Democratic Party( Slovenska Demokratska Stranka), led by Janez Janša, received around 27.9-28.1% of the vote. The minimal difference between the two parties underscores a pronounced degree of political polarization within Slovenian society. The electoral campaign was further complicated by serious allegations of foreign interference, involving the Israeli private intelligence firm «Black Cube», which Slovenian authorities have associated with alleged influence operations purportedly benefiting the opposition. These developments have raised significant concerns regarding the integrity of the electoral process, the resilience of democratic institution and broader issues of national sovereignty.   Election campaign and key political actors More than 10 political parties and coalitions participated in the parliamentary elections in Slovenia, with over 1300 candidates representing a wide range of political forces were nominated. Prime Minister Golob’s main political opponent, populist politician and supporter of US President Donald Trump, Janes Janša, previously served as Prime Minister of Slovenia three times. Golob, in turn entered national politics relatively recently, rising to power in 2022 as the leader of a center-left  coalition composed of three parties, making a significant shift in Slovenia’s political landscape. The election campaign of Janez Janša was structured around a clear confrontational logic, positioning the party in opposition to what it described as the «post-socialist left». A central element of the campaign strategy was the mobilization of voters through value-based polarization and competing visions of socio-economic development. The SDS presented itself as a political force aiming to «liberate the Slovenian people from the dominance of Golob’s government». In contrast, the governing center-left coalition led by Robert Golob was framed by its opponents as promoting policies associated with redistribution, egalitarianism and increased state dependence among citizens. In this context, Janša campaign adopted a distinctly adversarial rhetoric, encapsulated in the binary framing of «Us versus them», which reinforced the broader dynamics of political polarization within Slovenia’s electoral landscape.  The main lines of opposition in the SDP campaign emphasized «ambition and meritocracy instead of envy and egalitarianism». Janez Janša framed his political message around the idea of restoring motivation and purpose, arguing that societal progress depends on competition, performance-based rewards and a strong incentive for success as a means of improving the overall welfare of Slovenians. The party also promoted the conception of an «effective state», which prioritizes individual freedom and autonomy over dependence on government structures. This approach included calls for reduced bureaucracy, lower taxation and the creation of a more favorable environment for both citizens and entrepreneurs. Furthermore, the SDP strongly criticized what it described as the government’s migration policy, particularly its perceived lack of distinction between legal and illegal migration. Within this narrative, the party argued that the broader objective of the left was to alter Slovenia’s ethnic and cultural composition as well as the structure of the electorate, in order to secure long-term electoral advantages. On the other hand, the Freedom Movement (Gibanje Svoboda, GS) promoted a program centered on balanced and sustainable development, emphasizing a «green» transition, including plans related to the gradual phase-out of coal-based energy production such as the Šoštanj Thermal Power Plant (TEŠ 6), accompanied by socially just transition measures. The party’s agenda also highlighted innovation-driven economic growth, healthcare system reform and a pro-European, liberal-progressive political orientation. In the social policy domain, particular attention was given to improving quality of life, strengthening support for both younger and older generations and reducing social inequalities. The party sought to present a model of governance that reconciles environmental sustainability with economic competitiveness. Overall, the central message of the campaign was the pursuit of long-term sustainable development across social, environmental and economic dimensions, while reaffirming Slovenia’s role as an active and responsible member of the European Union and the broader European political community. On the economic front, the party set an ambitious objective of positioning Slovenia among the world’s 20 most competitive economies by 2030, accompanied by targeted annual GDP growth of approximately 3-5%. Particular emphasis is placed on strengthening innovation capacity, accelerating the transition to renewable energy sources and advancing digital transformation across key sectors of the economy. In addition, the program prioritizes the modernization of public service, especially in healthcare and education, as central pillars for long-term productivity and sustainable development. The party emphasizes a strong pro-European orientation and alignment of its program with the European Union’s strategic objectives, particularly in the areas of the green transition and climate neutrality. A key priority is promotion of a just decarbonization process (pravično razogljičenje), which includes the gradual phase-out of the TEŠ 6 coal-fired unit by 2030, accompanied by targeted  measures to support workers and regions economically dependent on the coal industry. Both major political forces have devoted significant attention to the future of the Šoštanj Thermal Plant (TEŠ 6), framing it as a central issue in debates over energy, economic policy and national sovereignty. The SDP has strongly criticized the government’s plan to phase out coal-based energy production, arguing that the premature closure of TEŠ 6 could undermine Slovenia’s energy independence and lead to substantial job losses, particularly in regions dependent on the coal industry. In contrast, Golob has justified the planned closure primarily on economic and regulatory grounds. He points to the increasing costs associated with the European Union Emissions Trading System (EU ETS) which has significantly raised the price of carbon allowance and consequently, the cost of coal-based electricity generation. With annual CO₂ emissions estimated at approximately 3.1-3.4 million tonnes at full capacity, TEŠ 6 is particularly exposed to rising carbon prices, which in 2026 fluctuated between €72 and €92 per tonne. As a result, the plant has become structurally unprofitable, promoting the government to allocate more than  €400 million in subsidies since 2024 to sustain its operations. Within this framework, the Golob government presents EU ETS obligations as a key economic rationale for the accelerated phase-out of TEŠ 6. Conversely, SDP portray such policies as externally imposed pressures from Brussels, framing the closure as a form of deindustrialization that threatens Slovenia’s energy security, employment, especially in the Šaleška Valley and the long-term stability of the country’s industrial base. The Šaleška Valley region ( including the towns of Velenje and Šoštanj), where TEŠ 6 is located has long been a center of industrial and mining activity, providing employment to several thousand workers directly and indirectly linked to the plant. The prospective closure of the facility therefor entails significant socio-economic risks, including job losses, declining regional incomes and potential increase in electricity and district heating prices for local residents. This issue became a central theme in the campaign of the SDP. Janša repeatedly framed the rapid phase-out of TEŠ 6 as «betrayal of workers» and an act of «energy capitulation», using the issue to mobilize support in industrial regions. In areas with strong ties to the energy and mining sectors, particularly in eastern and northeastern parts of Slovenian support for the SDP has traditionally been higher. In these regions, a considerable segment of the electorate perceived the policies of the «Freedom Movement» as a direct threat to their economic security, contributing  to stronger electoral backing for Janša despite Golob’s broader national appeal. Thus, what initially appears as a technical question of climate and energy policy has evolved into a deeper ideological confrontation between proponents of a progressive «green transition» and those prioritizing national energy sovereignty, industrial stability and economic pragmatism.   Election results and their consequences Pre-election polls and analytical forecasts consistently pointed to a highly competitive race, although at various stages of the campaign the momentum appeared to shift in favor of the opposition. In early 20266, the SDP gradually took the lead with several polls indicating support levels ranging from 20% to 28% at times placing it ahead of the ruling party. By February, the gap between the two main political forces had widened to approximately 5% points in favor of the SDP, raising expectations of a potential return of Janez Janša to power. By march, however, the situation had become increasingly uncertain. Aggregated polling data suggested that while the SDP maintained a slight lead at around 29% liberal bloc led by Golob remained electorally competitive, largely due to its coalition-building potential. In the final days before the election, most analysts converged on the assessment that the margin between the two camps would be minimal, likely resulting in a fragmented legislature without a decisive majority. In this context, although the opposition was at times perceived as the frontrunner, the final outcome ultimately confirmed the high volatility of the electorate and the deep polarization of Slovenian society.  The election was marked by high tensions around issues of migration, security and national identity. The «Black Cube» scandal has become a turning point in recent weeks, allegations have surfaced that representatives of an Israeli private intelligence company visited Slovenia, met with individuals associated with the SDP and were involved in leaking recording aimed at discrediting the Golob government on corruption charges. After the news became public, public opinion shifted sharply, support for the SDP declined, while the «Freedom Movement» began to gain momentum. As a result, Golob’s party previously trailing in the polls managed to secure a narrow electoral victory, winning by approximately 0.4-0.6%. The results of the Slovenian elections indicate the emergence of a fragmented political landscape in which no major party commands enough support to govern independently. Under these conditions, coalition-building with smaller parties such as the New Slovenia-Christian Democrats, the Slovenian People’s Party, the Focus Party, the Social Democrats and the Left Party will play a crucial role. This not only complicates the government formation process but also  introduces a degree instability, as governance will rely heavily on fragile inter-party compromises. At the same time, the elections exposed a pronounced level of political polarization. Voting patterns reflect deep structural cleavages, including urban-rural divides, generational differences and conflicts rooted in competing value systems. Voter turnout remained notably high, particularly around issues of identity and security, which further intensified the divisive character of the campaign. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.

The constitutional reform of 2026 in Kazakhstan represents one of the most significant stages of the country’s political modernization since gaining independence. The reform is framed by the official authorities as the logical culmination of the transition from a super-presidential model of governance to a more balanced system referred to as a “Just Kazakhstan.” It aims to improve mechanisms of public administration, strengthen state institutions, and enhance their interaction with society in the context of contemporary global challenges. The new Constitution, adopted through a national referendum on March 15, 2026, affects approximately 84% of the text of the previous 1995 Constitution (as amended in 2022). The document is set to enter into force on July 1, 2026. According to the official results published by the Central Referendum Commission, the draft was supported by 7,954,667 voters (87.15%). Voter turnout reached 73.12% of registered voters. These figures rank among the highest in national voting processes in the history of independent Kazakhstan.   Procedure of Preparation, Adoption, and Key Institutional Changes The preparation of the reform began in autumn 2025 at the initiative of President Kassym-Jomart Tokayev. In his Address to the Nation, the head of state explicitly emphasized the need to develop a “new Fundamental Law that corresponds to the realities of the 21st century.” On January 21, 2026, by Presidential Decree No. 1157, a Constitutional Commission of approximately 30 members was established, including representatives of the Presidential Administration, members of Parliament, qualified legal experts, and scholars. The Commission was tasked with preparing the draft within a highly compressed timeframe. Already on February 11, 2026, the President signed a decree calling for a national referendum, and on February 12, the full draft text was officially published on the state portals “gov.kz” and “election.gov.kz” Thus, only 22 days elapsed between the creation of the Commission and the publication of the draft. Public discussion lasted approximately one month and took place primarily through online platforms, as well as regional meetings organized by local executive bodies. The then-existing bicameral Parliament did not introduce amendments to the draft; instead, the document was submitted to a nationwide vote in its entirety. The scale of the transformation is unprecedented for the post-Soviet period. The previous Constitution consisted of 9 chapters and 98 articles, whereas the new one includes 11 chapters and 96 articles. Approximately 84% of the text has been fully revised or replaced. The key institutional changes can be grouped as follows: First, the reform fundamentally restructures the legislative branch. Instead of a bicameral Parliament (Mazhilis and Senate), a unicameral body – the Kurultai has been established, consisting of 145 deputies elected exclusively through a proportional party-list system. The abolition of the Senate and the concentration of legislative functions within a single body are intended to increase the efficiency of decision-making and eliminate potential delays in the legislative process. Second, the powers of the President in forming key state institutions have been expanded. Under the revised Articles 40–42, the President is granted the authority to appoint the Prosecutor General, the Chairs of the Constitutional and Supreme Courts, the Governor of the National Bank, and the Head of the National Security Committee. The Kurultai is informed of these appointments ex post facto. Third, the position of Vice President, absent in previous constitutional versions, has been reinstated. The Vice President is appointed by the President with the consent of the Kurultai by a simple majority vote and performs duties assigned by the head of state. Fourth, a new body – the Halyk Kenesi (People’s Council) has been established as a consultative-legislative organ under the President. A portion of its members is directly appointed by the President. This institution is intended to provide an additional channel for continuous dialogue between the state and society. Fifth, new provisions have been introduced in Chapter 2 on the rights and freedoms of citizens. Article 29 enshrines digital rights (including the protection of personal data in the digital environment), environmental guarantees have been strengthened, and Article 27 clarifies the definition of marriage as a voluntary union between a man and a woman. Additionally, a provision has been introduced prohibiting foreign funding of political parties and non-governmental organizations engaged in political activities (Article 23).   Declared Objectives of the Reform and Assessments by International Experts The official position of the Kazakh authorities presents the new Constitution as a document of a “mature state.” The key declared objectives of the reform include the redistribution of powers among branches of government according to the formula “a strong President – an influential Parliament – an accountable Government,” strengthening the system of checks and balances, promoting a human-centered approach, protecting traditional values, as well as ensuring digital and environmental rights. However, international experts offer a more cautious and неоднозначную assessment. Amnesty International characterizes the draft as a “concerning rollback in human rights and the rule of law,” pointing to the concentration of power in the presidency and restrictions on freedom of expression, assembly, and association due to the ban on foreign funding. Human Rights Watchemphasizes the weakening of checks and balances, the rushed preparation process, and the risks of using the reform to suppress dissent. According to a number of regional experts, these assessments are to a significant extent general in nature and do not fully take into account the specific political and social context of Kazakhstan as a Central Asian state. Citizens of Kazakhstan expressed broad support for the reform, indicating a high level of public approval. More than 10,000 proposals from citizens were submitted through the platform “referendum2026.kz” highlighting the popularity of the document.   Conclusion The constitutional reform in Kazakhstan represents a significant stage in political modernization and has contributed to the creation of a more compact and efficient system of public administration. The declared objectives of strengthening institutions, promoting a human-centered approach, and building a “Just Kazakhstan” received broad public support during the referendum. This experience may serve as a practical example of balancing the efficiency of state institutions with maintaining control over political continuity. Certain mechanisms introduced by the reform – such as the institution of the Vice President, the unicameral legislature, and restrictions on foreign funding – may be adaptable to other governance models, potentially contributing to long-term political stability. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.

By Ilyos Shaymardonov, UWED Master’s Student, IAIS Research Intern   Introduction.The outbreak of the US-Iran war in early 2026 shook the global geopolitical architecture and put the strategies of major powers to a direct test. In the view of many experts, this conflict appeared to open the door to ready-made strategic opportunities for China. It was assumed that while the United States was entangled in another costly military campaign in the Middle East, Beijing would seek to fill the resulting vacuum, deepen its ties with Tehran, and accelerate its challenge to American leadership. However, this hypothesis misinterprets how China actually behaves amidst geopolitical instability. An analysis of the situation reveals that China’s response to the US-Iran war demonstrates not expansionism, but a doctrine of calculated restraint. This, in turn, is a deliberate position stemming from China’s firm conviction regarding global economic vulnerabilities, its long-term investments within the framework of the «Belt and Road» initiative, and the belief that systemic stability, not short-term tactical gains, is the primary condition for its sustainable development.   The foundations of the China-Iran strategic partnership: The «Belt and Road» and a 25-year agreement To understand China’s position during the war, one must first understand the nature and depth of the pre-war China-Iran relations. In March 2021, China and Iran signed a 25-year Comprehensive Cooperation Agreement. According to him, China has committed to investing up to $400 billion in the country’s infrastructure and energy sector in exchange for stable supplies of Iranian cheap oil. This agreement directly linked Iran to the «One Belt, One Road» initiative, transforming Tehran into a crucial hub connecting Central Asia with the Persian Gulf, and this corridor would allow China to bypass strategic maritime chokepoints dominated by the U.S. Navy. Iran’s official accession to the Shanghai Cooperation Organization in 2023 further solidified this closeness in a formal-institutional sense within the network of regional institutions led by China. The strategic logic underlying this partnership is multifaceted. As noted by Iranian expert S. Madani, the «Belt and Road» initiative is not merely an infrastructure program but also serves as an economic tool designed to expand its sphere of influence and acquire resources amidst systemic constraints. For Iran, this initiative serves as a mechanism to maintain institutional stability and bypass sanctions, providing access to markets and capital that are difficult to attain under Western pressure. In this way, China became the primary buyer of Iranian oil, purchasing approximately 90% of the country’s crude exports. The Hudson Institute describes this relationship as crucial for Iran’s economic survival and central to Beijing’s concept of creating a secure land-based energy corridor free from U.S. maritime dominance. From a geopolitical standpoint, this cooperation served China’s overarching interest in establishing a multipolar world order. Iran’s strategic geographic location connecting Central Asia, the Middle East, and the Persian Gulf has given China a diplomatic edge in a region where U.S. influence has historically been dominant. Analysts at the NESA Center observe that China’s geopolitical expansion into the Middle East is being carried out incrementally and in a risk-averse manner to avoid provoking a sharp reaction from regional or international powers. In this process, cooperation with Iran serves as a vehicle for strengthening China’s position in the Middle East without engaging in direct conflict.   Calculated restraint: why is Beijing not joining the war? Despite the depth of Sino-Iranian relations, China’s reaction to a potential US-Iran war has been notably restrained. Beijing lacks a permanent military presence in the Greater Middle East, has no security commitments to Tehran, and does not possess the real capability to shift the military balance in such an intense conflict. Deploying an aircraft carrier group to the region to protect Chinese oil tankers would be an extremely dangerous risk for a navy that is only just beginning to develop its open-sea capabilities. This military restraint is not an admission of weakness but rather a rational response to the U.S.-China rivalry. Analysts from «Foreign Affairs» note that Chinese leaders do not view every U.S. failure as a Chinese success, nor do they believe it is necessary to seize every geopolitical opportunity. Beijing’s calculus will be determined by whether the situation stabilizes or, conversely, descends into chaos. A regional conflict centered on the Strait of Hormuz is, in turn, a source of such chaos. This approach contrasts with the «win-lose» logic ingrained in Washington’s strategic debates. While American analysts view China as the biggest beneficiary of America’s overextension, Beijing sees a destabilizing crisis that threatens the open trading environment which is the foundation of its economic model. China’s diplomacy is meticulously crafted to reflect this very position. While Chinese Foreign Minister Wang Yi called for maintaining the diplomatic momentum in the negotiations, Xi Jinping openly criticized the US blockade of Iranian ports as a dangerous and irresponsible act and leveraged his ties with Pakistan to urge Iranian negotiators to soften their stance. The Atlantic Council notes that this two-pronged policy-namely, openly condemning U.S. coercive measures while covertly applying diplomatic pressure on Tehran – clearly shows Beijing’s aspiration to present itself as a responsible mediator rather than a sponsor of the war.   Economic vulnerabilities and tensions in the strait of Hormuz The most significant systemic challenge China faces in this conflict is in the energy sector. China is the world’s largest oil importer, with about 70% of the country’s needs met by foreign sources, of which approximately one-third was previously transported through the Strait of Hormuz. The de facto closure of this strategic transit corridor since late February 2026 has dealt a severe blow to China’s energy security. However, Beijing is not panicking, and this composure is the result of many years of meticulous preparation. China had been building up large crude oil reserves since 2025, a time when an oversupply in the global market and low oil prices created a unique opportunity to accumulate reserves at a low cost. Beijing is also consistently pursuing a policy of diversifying its supply sources. Following the restoration of ties temporarily severed by U.S. sanctions, China’s national oil companies have resumed purchasing Russian oil by sea. Meanwhile, the «Power of Siberia 1» pipeline continues to deliver Russian gas in volumes exceeding its designated capacity.  According to an analysis by the Stimson Center, the war has accelerated a simultaneous two-way approach: purchasing from different sources and investing in green energy. Specifically, Beijing is expanding China-Algeria energy agreements through Sinopec and intensifying cooperation in clean energy with Morocco and Egypt to reduce long-term dependence on the Persian Gulf. Disruptions at Qatar’s Ras Laffan liquefied natural gas facility have led to a declaration of force majeure on contracts with Chinese buyers, starkly highlighting the real economic costs Beijing is facing. The rising cost of energy, delays in cargo deliveries, additional insurance payments, and expenses related to rerouting shipping lanes are increasing production costs in China’s export-oriented industries. This, in turn, is causing a decline in global market demand at a time when Chinese manufacturers are facing pressure from excess capacity domestically. None of this aligns with Beijing’s strategic interests, which reinforces the key analytical conclusion that the war is not the geopolitical gift for China that Washington analysts sometimes suggest.   A diplomatic approach and a multipolar worldview The prevailing view emerging from the conflict – that the actions of the U.S. and Israel were an act of aggression, that Iran’s response was reactive in nature, and that this conflict clearly demonstrates the price of American hegemony – is nearly identical to Beijing’s long-standing criticism of U.S. global policy. China’s persistent call for multipolarity is taking on new meaning in the current context. The fact that the economies of more than 145 countries now trade more with China than with the U.S., combined with the serious precautions U.S. allies in Europe and Asia are taking due to energy supply disruptions, reinforces the credibility of a structural basis for a more fragmented international order. China’s vote against UN Security Council sanctions resolutions concerning Iran and the Middle East as early as March 2026, despite calling for Tehran to refrain from direct material support, will strengthen its reputation as a counterweight to Western institutional dominance. An analysis of the Belt and Road Initiative in the Middle East by the Andersen and Sending think tanks sheds light on this matter. These institutions have consistently portrayed China’s infrastructure investments as a strategic threat, often significantly exaggerating the true scale of Belt and Road projects in the region. However, it is precisely this approach that inadvertently validates China’s self-portrayal as a constructive alternative power. The war will accelerate this process, as it will provide clear proof for an argument long made by critics of U.S. hegemony. Namely, that America’s military adventures create such instability that the existing order cannot absorb it without suffering losses.   Limits and long-term implications of China’s strategy China’s strategy of calculated restraint is not without its own weaknesses and contradictions. The most prominent of these is the problem of «dual-use technologies. According to reports from the Foundation for Defense of Democracies, China-affiliated firms provided commanders of the Islamic Revolutionary Guard Corps of Iran with high-resolution satellite imagery of U.S. military installations, including the Prince Sultan Air Base. Whether such activities are state-sanctioned or are the actions of commercial entities operating in a «gray area» this exposes Beijing to counter-pressure from Washington and undermines its position as a neutral diplomatic intermediary. The long-term strategic loss could be even more severe. China had relied on a self-confident, defiant Iran with nuclear ambitions as the cornerstone of its «Belt and Road» strategy in the Middle East. Having suffered a heavy blow and been caught in a whirlwind of internal conflict, Iran can no longer serve as a reliable factor to distract American attention. The «Belt and Road» projects, which are dependent on Iran’s ports, railway corridors, and energy infrastructure, are now facing even greater instability. The idea of establishing a secure land-based energy corridor that bypasses key maritime points dominated by the U.S. has been dealt a serious blow. Furthermore, China’s long-standing efforts to simultaneously support Iran while balancing relations with the Arab monarchies of the Persian Gulf have been exposed as inherently contradictory. The Atlantic Council’s scenario analysis outlines four possible trajectories for the post-war geopolitical order, ranging from a limited U.S. success that maintains systemic stability to a scenario where China responds with significant indirect intervention. Such an intervention would include providing advanced intelligence and logistical support to Iranian forces, as well as applying coordinated pressure in parallel theaters like the Taiwan Strait, thereby transforming the regional conflict into a systemic confrontation. Beijing will definitely prefer the first direction. Any escalation in the second direction would mean a defeat, not a victory, of the Chinese strategy.   Conclusion. China’s role in the geopolitics of the US-Iran war is neither consistent with the statement that «Beijing is the main beneficiary of the war» nor with the dismissive statement that «China is simply not in this process». Evidence suggests that Beijing is following a well-thought-out doctrine of restraint. It bears real economic costs and skillfully manages structural vulnerabilities in the energy sector. It is using the diplomatic fallout from American military action to advance its own idea of a multipolar world and is conducting covert pressure through channels that preserve the possibility of rejection. This position reflects a strategic culture that sees systemic stability as a necessary condition for national power and sees America’s excessive expansion as a cautionary lesson rather than an opportunity to expand. This war, however, exposed the limitations of a grand strategy built on weak cooperation. A weakened Iran, disrupted Belt and Road corridors, and tight control over the transfer of dual-use technologies are all losses that Beijing did not choose and cannot easily compensate for. Whether China’s abandonment of tactical options in favor of strategic patience serves its long-term interests in the Middle East will depend on how and how quickly the war ends. The oversimplified «win-lose» pattern that prevails in Washington’s China-related debates is more of a blur than an illumination. In this crisis, China’s strength lies not in what it has done, but in what it has managed to discipline itself not to do. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.

The global economy is undergoing one of the largest infrastructural technological shifts since the Industrial Revolution, driven by the formation of a global artificial intelligence infrastructure. The global economic architecture has entered a qualitatively new phase of technological development, the key feature of which is the transition from a predominantly software-based model of the digital economy to a capital-intensive infrastructural model of artificial intelligence (AI). While in previous decades software products and intellectual property were the main drivers of value growth for technology companies, the material computing infrastructure is now becoming the foundation of economic dynamics. This involves the construction of large data centers, the procurement of specialized graphics processing units (GPUs), the creation of cooling systems, and the development of energy grids capable of supporting the operation of high-performance computing clusters. In effect, a new technological ecosystem is taking shape, where the physical provisioning of computing power is acquiring strategic importance comparable to the role of transportation infrastructure or energy systems during the industrial era. Statistics on the distribution of data centers by country provide insight into the actual configuration of the global AI infrastructure. As of early 2026, there are approximately 10,800large data centers worldwide, and their geographic distribution is highly uneven. The United States remains the absolute leader, where about 3,960 centers are located, accounting for approximately 37% of the global infrastructure. Thus, the US possesses nearly four out of every ten data centers globally, providing the country with a significant technological advantage in cloud computing and AI systems development. Significantly smaller, yet notable infrastructural clusters have been established in the United Kingdom (498), Germany (470), China (365), and France (335). Other major players include Japan (249), Australia (268), India (275), and Brazil (198). Notably, despite active state support, China currently lags behind several European countries in terms of the number of data centers. The geographical structure of this infrastructure indicates a high concentration of computing power in the developed economies of North America, Europe, and East Asia, while a significant portion of countries in Asia, Africa, and Latin America possesses only limited capabilities for hosting such facilities. This asymmetry creates a persistent gap in access to computing resources and increases the dependence of many nations on the infrastructure of leading technological hubs. The scale of investment in AI infrastructure has already reached levels comparable to the largest industrial projects in history. According to available estimates, the aggregate capital expenditures of the four largest US tech corporations—Microsoft, Alphabet, Meta, and Amazon—amounted to between $370 billion and $700 billion by the end of 2025. If current growth rates persist, their investments could exceed 2% of the US GDP. For historical comparison, a similar scale of capital investment was observed during the construction of the US national railway network in the 19th century, which laid the foundation for the country’s industrial leadership for decades to come. Unlike those projects, however, the current technological infrastructure is being built primarily by private capital and with a significantly higher degree of uncertainty regarding future economic returns. At the same time, concerns are increasingly voiced within the expert community regarding the possible formation of an infrastructural bubble in the computing power market. Major tech companies are building massive data centers and purchasing millions of specialized processors based on expectations that new generations of AI models will drive a sharp increase in labor productivity and create fundamentally new markets for digital services. However, the actual scale of commercial demand for such capacities remains uncertain. If the anticipated economic impact of AI adoption falls short of projections, a significant portion of the built infrastructure may end up underutilized. Unlike traditional infrastructure assets, such as transportation systems or energy grids, data center equipment has a relatively short technological lifecycle. Specialized chips become obsolete within just three to four years, requiring constant upgrades and additional investments. Consequently, the risk of accumulating unneeded computing power is compounded by the rapid rate of technological obsolescence of the equipment, making the current industry development model particularly sensitive to demand fluctuations. A prime example of mounting financial risks is the situation surrounding the American corporation Oracle. The company, long considered one of the key players in enterprise software and cloud solutions, found itself under significant debt pressure by the spring of this year. According to various estimates, its total debt obligations exceeded $137 billion. In a high-interest-rate environment, such a debt burden severely limits the company’s ability to further expand its infrastructure projects. Major rating agencies already warn about the likelihood of downgrading the company’s credit rating to the high-risk category. Attempts to raise additional funding through bond and stock issuances provoked a negative market reaction, resulting in the company’s stock price more than halving since the fall of 2025. Corporate circles are discussing possible measures for massive staff reductions and the revision of investment plans. A particular vulnerability of the current financial structure is the interdependence between large tech companies and AI developers. For instance, the massive Stargate infrastructure project, aimed at developing computing power for the leading developer OpenAI, entails Oracle’s participation as a major cloud infrastructure provider. Meanwhile, OpenAI itself remains an unprofitable company, with projected operating losses that could amount to approximately $14 billion in 2026. A complex system of mutual investments and financial flows has formed within the industry, where funds invested by some tech companies into AI developers are subsequently channeled to pay for the cloud services of other companies, which, in turn, use the acquired resources to purchase equipment from processor manufacturers. Such a closed-loop financing structure amplifies systemic risks, as a failure in any single element of the chain could trigger a chain reaction in a market valued at trillions of dollars. In addition to financial constraints, the development of AI infrastructure faces severe energy barriers. Modern data centers require massive amounts of electricity, and their consumption continues to grow rapidly. According to estimates by American think tanks, in the coming years, data centers could consume up to 17% of all electricity generated in the US. This load puts significant strain on existing power grids, many of which were built decades ago and are not designed for such intensive consumption. In a number of regions, including the states of Virginia and Texas, action plans are being discussed in the event of grid overloads and power outages. In response to this situation, US federal regulatory agencies have initiated stricter rules for connecting large data centers to the national power grid. The new regulatory policy effectively requires tech corporations to independently invest in the construction of generating capacities and the modernization of regional energy infrastructure. This has led to the emergence of a new trend, dubbed the tech-led "nuclear renaissance" within the expert community. Microsoft initiated a project to modernize and restart a reactor at the Three Mile Island nuclear plant, costing approximately $1.6 billion. Amazon acquired exclusive rights to use the power output of the Susquehanna nuclear power plant, while Google signed contracts for the development and deployment of small modular reactors (SMRs). Thus, tech corporations are gradually transforming not only into digital infrastructure operators but also into major investors in the energy sector. Nvidia continues to play a pivotal role in the new technological ecosystem, having effectively become the monopoly supplier of specialized GPUs for AI systems. The company’s market capitalization is approaching $3 trillion, with up to 90% of its revenue coming from the data center segment. However, this position is also facing new challenges. Tighter US export restrictions have limited the supply of specialized processors to the Chinese market and have already led to multibillion-dollar write-offs. Simultaneously, the technological rivalry between the US and China is intensifying. China is implementing an alternative model for computing infrastructure development based on centralized state planning. The national "Eastern Data, Western Computing" program envisions locating large computing centers in regions with affordable electricity and favorable climatic conditions. This allows for a significant increase in data center energy efficiency and a reduction in operational costs. According to some estimates, China’s technological hardware lag behind the US could narrow to a matter of months. At the same time, the Chinese AI operational model entails significant resources being directed toward internal control, data verification, and compliance with state information security standards, which reduces the efficiency of computing power utilization. An additional factor amplifying the strategic importance of AI infrastructure is its rapidly growing role in the military domain. In recent years, leading powers have increasingly viewed computing power and machine learning algorithms as key elements of future military superiority. Unlike previous technological revolutions, where individual types of weapons played a decisive role, the current phase is characterized by the integration of AI into decision-making, reconnaissance, and operations management systems. This means that control over computing infrastructure is gradually turning into a critical factor for national security and military balance. As of today, one of the most prominent examples of AI application in combat conditions has been the military conflict in the Middle East. During the military campaign against Iran, US military structures have been actively utilizing Project Maven, a system developed by Palantir. The system is designed to process massive arrays of intelligence data coming from satellites, unmanned aerial vehicles (UAVs), and other sensor platforms. It automatically identifies potential targets, ranks them by priority, and generates recommendations for operators. The system also integrates Claude, a large language model by Anthropic, which is used to process intelligence reports and accelerate information analysis. According to military experts, the application of such algorithms has allowed for a sharp increase in the speed of operations: in the first 24 hours of the campaign alone, over a thousand targets were identified and struck in the area of operations, and by mid-March, that number had exceeded three thousand. The algorithms are capable of generating up to a thousand target recommendations per hour, which significantly exceeds the capabilities of traditional human analysis. This experience has revealed both the advantages and limitations of such technologies. Proponents of AI application point to increased strike accuracy and the reduction of human error in intelligence analysis; however, the effectiveness of such systems decreases in complex weather conditions and when decoys are used. Furthermore, the use of civilian AI models in military operations has already sparked legal and political disputes between defense departments and tech companies. Nevertheless, military leadership views AI as a crucial tool for reconnaissance and targeting, and the ongoing campaign in the Middle East has become one of the most large-scale tests of these technologies. In a broader context, this indicates that the rivalry between the US and China in the sphere of computing power is acquiring a military-strategic dimension. Amid the rapid growth of computing capabilities, international discussions on regulating the sector are intensifying, including limits on the volume of computing resources held by private companies and the licensing of high-performance GPU usage. Until recently, this idea was considered fringe, but today it is gradually entering the agenda of intergovernmental consultations. The formation of the global AI infrastructure is becoming one of the key economic processes of the decade. On the one hand, it unlocks new sources of growth; on the other hand, massive investments and energy constraints create significant systemic risks. Strategically, this signals the formation of a new foundation for the global balance of power. While in the 20th century the key factors of influence were industrial potential, energy resources, and nuclear technology, today these are joined by a state’s ability to provision massive computing capacities and control AI infrastructure. This makes the possession of a sovereign computing base one of the core conditions for maintaining economic and technological independence in the long term. * The Institute for Advanced International Studies (IAIS) does not take institutional positions on any issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of the IAIS.