Between Washington and Beijing: Can Central Asia Preserve Its Multi-Vector Technology Policy?

The United States is preparing to tell dozens of partner countries that they may have to choose between American and Chinese approaches to cooperation in artificial intelligence. In effect, Washington is asking its allies and partners to take a clearer position as global technological competition between the United States and China intensifies.

According to a draft letter from the U.S. State Department seen by Reuters, the issue concerns 35 countries that signed the American AI Opportunity Statement adopted in June. The draft letter warns that these countries may not be able to participate simultaneously in the U.S.-led Pax Silica initiative and a competing Chinese initiative on artificial intelligence.

As AI becomes increasingly connected to semiconductor manufacturing, critical minerals, energy, data centers, computing capacity, and national security, Central Asian countries will face growing pressure over their choice of technology partners.

1. Is it realistic to expect Central Asian countries to choose between American and Chinese AI ecosystems?

It is realistic to expect Washington to become increasingly demanding about where its partners stand on certain sensitive technologies. It is much less realistic, however, to expect Central Asian countries to abandon their multi-vector foreign policies and choose one technology ecosystem at the expense of the other.

This distinction is fundamental. The United States increasingly views artificial intelligence not simply as an economic sector but as a strategic technology ecosystem that includes advanced semiconductors, computing capacity, data centers, critical minerals, energy infrastructure, research networks, technology standards, and investment.

From Washington’s perspective, a country’s participation in China-linked initiatives such as WAICO may therefore raise questions that go far beyond AI regulation itself. The American argument is that technology partnerships can create long-term dependencies and may potentially facilitate the transfer, spread, or circumvention of restrictions on technologies that Washington considers strategically sensitive.

This is why the State Department draft letter is important, even though, according to available information, it has not yet been sent to its recipients and its contents may still change. The main message behind the initiative appears to be that Washington sees participation in Pax Silica not simply as membership in another international forum but as a strategic commitment.

WAICO and Pax Silica are not mirror-image organizations: one important point should be kept in mind: WAICO and Pax Silica should not be treated as fully symmetrical organizations. Pax Silica is primarily focused on the security and resilience of critical technology supply chains—from minerals and energy to semiconductors, computing infrastructure, and artificial intelligence.

Based on available information, WAICO places much greater emphasis on building national capacity, international AI governance, access to technology, workforce development, and greater participation by developing countries in the emerging global AI ecosystem.

It may therefore be more accurate to describe them not as two identical “AI blocs,” but as two competing approaches to the emerging technological order. The American approach can be broadly described as “Secure the technology ecosystem.” The Chinese approach is more focused on another idea: “Expand access to the technology ecosystem.”

For countries in the Global South, the second approach is particularly attractive. This does not mean that China is acting purely out of altruism. By providing infrastructure, equipment, AI models, training, and technology platforms, Beijing gains clear strategic advantages. The wider use of Chinese technologies also helps spread Chinese standards, products, and institutional practices.

Independent research on WAICO highlights precisely this dual nature. On the one hand, the organization may strengthen China’s international influence. On the other, it responds to a genuine demand from developing countries for training, infrastructure, access to open-weight models, and a greater role in shaping international AI rules.

At the same time, the AI race is entering a new phase. China is no longer simply catching up with the United States. It is competing with the U.S. in performance, cost, and scale of deployment. According to Stanford HAI, in March 2026 the gap between the leading American and Chinese AI models was only 2.7%. Since the beginning of 2025, models from the two countries have repeatedly exchanged the lead.

China is strengthening its position through DeepSeek, Kimi K3, Qwen, GLM, and MiniMax—open-weight models, rapid development, and significantly lower operating costs. Price may become one of the most important factors in the next stage of the AI race. The US still has major advantages: advanced chips and computing infrastructure, capital, and leading AI research laboratories. In 2025, U.S. developers produced 59 notable AI models, compared with 35 in China.

But China’s strategy is different: Open models + low cost + industrial scale + government support + rapid deployment. The key question is therefore no longer simply who has the most powerful AI. It is increasingly about who can make advanced AI cheap, accessible, and widely available.

That could determine not only technological leadership but also geopolitical influence in the global AI race. For Central Asia, this distinction is particularly important. Central Asia: Diversification, Not Choosing Sides. The foreign policy model of Kazakhstan, Uzbekistan, and other Central Asian countries has traditionally been based on diversification rather than exclusive alignment with a single external partner.

Their goal is not so much to balance ideologically between the United States and China as to avoid excessive dependence on any one external power while attracting investment, technology, infrastructure, and access to different markets.

This approach is particularly rational in the technology sector. Central Asian countries do not have sufficient resources to build a complete modern AI ecosystem entirely on their own.

Building competitive AI infrastructure requires not only skilled professionals but also substantial financial investment, access to advanced technologies, and integration into global technology chains. For most Central Asian countries, building such an ecosystem independently would be extremely expensive and technologically difficult.

The region therefore depends on foreign capital, advanced GPUs and computing capacity, cloud platforms, telecommunications equipment, and investment in data centers. International research partnerships, training for engineers and tech specialists, and access to global markets and tech platforms are equally important.

In other words, the key question for Central Asia is not whether the region can build its own AI ecosystem completely from scratch. The real question is whether it can integrate effectively into the global AI infrastructure while attracting the technology, capital, and expertise it needs. That is why international technology and investment cooperation is not simply an additional advantage. It is a necessary condition for building a competitive AI economy in Central Asia.

The economic cost of cutting cooperation with one of the world’s two largest technology centers could therefore be very high. At the same time, Central Asia possesses resources that both sides need. The region has strategic importance because of its energy resources, critical minerals, geographic position, growing digital infrastructure, and emerging markets.

This gives Central Asian countries some room for maneuver. They should not be seen simply as passive recipients of American or Chinese technologies. Competition among external partners can be used to negotiate better investment terms, local production, and technology transfer.

For example, Kazakhstan’s simultaneous engagement with American and Chinese technology and investment partners shows that Astana views technology diversification as part of its national strategy. The same logic applies to a large extent to Uzbekistan, although its economic structure and external partnerships have their own specific features.

For this reason, I would not say that Central Asian countries are already facing an immediate need to “choose sides.” Rather, we are entering a period in which the cost of maintaining genuine technological multi-vectorism will increase.

Where Will Multi-Vectorism Face Limits?

The main problem is that U.S.-China competition is gradually creating two technology ecosystems that may become incompatible in the most sensitive areas. A country can use Chinese technologies in commercial and civilian applications while cooperating with the United States on initiatives aimed at strengthening the resilience of global supply chains.

However, the room for such technological balancing becomes much smaller when it comes to advanced GPUs, dual-use technologies, military AI applications, sensitive government data, high-performance computing, and advanced semiconductor manufacturing. Infrastructure directly affected by U.S. export controls is particularly important. In these areas, Washington is likely to demand greater transparency and increasingly seek restrictions on the use of certain Chinese technologies and suppliers.

At the same time, Beijing may be less comfortable with a situation in which a country actively uses Chinese technology infrastructure while participating in initiatives designed to reduce China’s role in strategic technology supply chains. This creates a new strategic dilemma for countries that want to cooperate with both the United States and China. Technological neutrality is becoming increasingly difficult precisely where AI, semiconductors, data, and computing capacity intersect with national security.

As a result, choosing a technology partner may gradually stop being purely an economic decision and become a question of foreign policy and strategic positioning. A new type of pressure is therefore emerging. The question may no longer simply be, “Are you choosing America or China?” Instead, it may become, “Which parts of your technology ecosystem do you consider secure and trusted?” That is a much more serious question.

For Central Asian governments, the optimal response would be to preserve technological diversification while developing clearer national rules for sensitive technologies. It is important to distinguish ordinary commercial cooperation from strategically sensitive infrastructure rather than treating all Chinese or American technologies as if they were the same.

In other words, Central Asia needs to preserve strategic flexibility while avoiding technological uncertainty in areas where genuine national security concerns exist.

The Risks of an “Either-Or” Policy

There is also a broader diplomatic dimension. If Washington begins approaching Central Asian countries with an openly bipolar Cold War-style formula—”either you are with us or against us”—the result could be the opposite of what the United States wants.

Governments in the region may see such pressure as an attempt to limit their sovereignty. As a result, they may become less willing to move closer to Washington and, instead, gain additional reasons to deepen their relationships with China and other partners.

The United States should therefore focus not only on restrictions but also on creating an attractive alternative. If Washington wants Central Asian countries to reduce their dependence on Chinese technology, political arguments alone will not be enough. The region needs a real and competitive alternative.

That alternative should include affordable financing, infrastructure projects, modern computing capacity, workforce training, technology cooperation and technology transfer, access to international markets, and long-term investment.

For Central Asian countries, what matters most is not political declarations but concrete economic and technological results: Who is ready to build data centers? Who will provide computing capacity? Who will invest in digital infrastructure? Who will train specialists? Who will help local companies reach global markets?

In the long term, therefore, the most effective tool of geopolitical influence in Central Asia will not be a demand for political loyalty but the quality of the technological and economic alternative being offered.

In other words, if Washington wants to compete with China for influence in the region, it should focus less on persuading Central Asian countries to abandon Chinese technology and more on creating conditions in which alternative partnerships are objectively more attractive, affordable, and promising.

2. What Does This Mean for Uzbekistan Given Its Growing Economic Ties with China?

This issue is particularly important for Uzbekistan because its relationship with China has long gone beyond traditional trade. China is one of Uzbekistan’s most important economic partners. Chinese companies are increasingly involved in infrastructure, industry, energy, transport, telecommunications, and digital projects. Bilateral trade is growing rapidly, while Chinese investment and the number of companies with Chinese capital are also increasing.

According to official data, bilateral trade with China reached almost $18 billion in 2025, while Chinese investment in Uzbekistan increased fivefold to approximately $17 billion. More than 6,000 companies with Chinese capital operate in the country.

But an even more important development is taking place: relations between the two countries are gradually moving from trade and investment toward technological cooperation.

In April 2026, Uzbekistan and China agreed to expand cooperation in AI, digital technologies, new materials, green energy, and research infrastructure. The agenda includes technology transfer, commercialization of research, joint investment funds, technology parks, and exchanges between researchers.

This means that any potential U.S. demand that Uzbekistan “choose” between competing technology ecosystems would not be an abstract question of foreign policy alignment. It could directly affect specific investment projects, technology partnerships, and infrastructure already connected to China.

Partnership with China doesn’t mean a geopolitical choice in China’s favor

It is important to emphasize that Uzbekistan’s growing ties with China do not automatically mean a geopolitical choice in favor of Beijing. Tashkent has consistently pursued a diversified foreign economic policy.

China remains a key partner, but Uzbekistan is also seeking to develop economic and technological ties with the United States, the European Union, South Korea, Japan, Türkiye, the Gulf Cooperation Council countries, and international financial institutions.

This is particularly visible in the technology sector. Uzbekistan needs multiple sources of investment and expertise because its goal is not simply to import ready-made technologies but to build its own digital and industrial capabilities.

That is why Tashkent is interested in cooperation with the United States in AI and advanced technologies while also seeking to maintain diversified partnerships and openness to mutually beneficial international cooperation.

Why Tashkent Is Unlikely to Accept a Hard “Either-Or” Choice

For Uzbekistan, balanced multi-vector diplomacy is not only a matter of foreign policy positioning. It is also an important economic strategy. The country needs Chinese investment and technology, American capital and advanced technologies, European markets, and South Korean and Japanese technologies, as well as deeper regional economic integration.

New transport corridors, energy and renewable energy projects, modern computing capacity, and digital infrastructure are equally important. All of these areas require substantial investment, technology, and access to international markets. In this environment, competition among external partners creates additional opportunities for Uzbekistan.

The more cooperation options the country has, the stronger its negotiating position and the greater its chances of securing better terms for investment, technology, infrastructure projects, and market access.

Therefore, voluntarily giving up one major source of investment, infrastructure, and technology simply to accommodate the geopolitical preferences of another major power would make little economic sense.

For Uzbekistan, the rational strategy is not to choose one external center of power but to use competition among them as effectively as possible while preserving room for independent economic and technological decisions. A more sensible strategy is to maximize the benefits of competition while reducing dependence on any individual partner.

If Washington genuinely wants to help Uzbekistan diversify its technology relationships, it will need to offer concrete alternatives: investment in data centers, advanced computing, AI research, cybersecurity, education, human capital, and semiconductor-related supply chains.

China is already offering Uzbekistan significant opportunities in infrastructure, economic connectivity, and technology. The United States and other Western partners could compete using their own strengths: advanced AI research, software, venture capital, high-performance computing, international standards, education, and access to Western markets.

For Uzbekistan, this could be a favorable situation—provided that competition among major powers does not turn into a constraint on Tashkent’s own political and economic autonomy.

Uzbekistan’s Technological Sovereignty

There is another reason why Uzbekistan needs to maintain access to both technology ecosystems. The country’s own AI strategy is still developing. Unlike the United States or China, Uzbekistan does not need to protect an already established national AI ecosystem from an external competitor. Its task is to build a competitive AI ecosystem of its own. That requires the widest possible access to knowledge, computing resources, investment, and human capital.

From a strategic perspective, it would therefore be unwise to voluntarily cut off access to Chinese AI technologies, particularly relatively affordable and open technologies. The main goal for Uzbekistan should not be isolation from individual technology partners but diversification.

At the same time, as the digital economy develops, increasing attention must be paid to ownership and control of critical digital infrastructure. It is essential to know who owns key data centers, where sensitive government data is stored, and who controls cloud infrastructure.

Other important questions include whether algorithms can be independently audited, how easily a foreign supplier can be replaced, and whether the country could face excessive costs when switching to another technology platform.

Another major risk is vendor dependence. If moving to another platform is prohibitively expensive, or if critical systems cannot be operated without foreign specialists, formal diversification of suppliers does not guarantee real technological sovereignty.

Uzbekistan therefore needs to develop its own capabilities and train specialists who can independently operate and maintain critical digital infrastructure. Ultimately, control over data, infrastructure, technology, and key expertise will determine the country’s level of digital sovereignty.

From this perspective, the formal question of whether Uzbekistan participates in WAICO or a particular U.S. initiative is less important than whether the country retains the ability to make its own technological decisions and change partners without suffering critical economic or infrastructure losses.

Digital sovereignty should not mean rejecting foreign technology. It should mean being able to use foreign technology without losing the ability to make strategic decisions independently.

3. How Is Uzbekistan Most Likely to Respond to a Direct U.S. Approach?

If Washington officially asks Uzbekistan to choose between American and Chinese approaches to AI cooperation, Tashkent is likely to respond cautiously and avoid accepting the idea that it must make a geopolitical choice.

The initial response would most likely be diplomatic rather than confrontational.

Uzbekistan would probably emphasize its sovereign right to develop international partnerships based on its national interests and to cooperate with different countries in different sectors.

At the same time, Tashkent would likely seek clarification from Washington about what exactly the United States considers incompatible. This is crucial. There is a major difference between requiring compliance with U.S. export restrictions on certain categories of advanced semiconductors and demanding that Uzbekistan completely end technological cooperation with China.

The first could potentially be negotiated. The second would be much harder to accept. Tashkent would therefore likely try to transform the political question—”Whose side are you on?”—into a technical and legal one:

“Which technologies are restricted? Which suppliers are covered? What security measures are required? What data may be transferred? Which infrastructure is considered sensitive? What alternative technologies and financing mechanisms are the United States prepared to offer?”

Such an approach would allow Uzbekistan to preserve diplomatic room for maneuver. Why Tashkent Has No Interest in Breaking with China. Uzbekistan has serious economic reasons to avoid a sharp deterioration in relations with China. This partnership has developed over many years and includes trade, investment, infrastructure, energy, manufacturing, and technology.

Trying to dismantle this network of relationships quickly would create significant economic costs. The possible reaction from Beijing must also be considered. China does not necessarily need to issue a public ultimatum. It has significant economic tools at its disposal.

Chinese companies are important suppliers of infrastructure and technology, while Beijing can use investment decisions, financing, market access, and technology partnerships to strengthen its position. For Uzbekistan, the safest strategy is therefore to avoid a situation in which maintaining relations with China comes into direct conflict with access to American technologies.

Most likely, Tashkent will seek a formula along the following lines: Uzbekistan is open to cooperation with the United States in advanced AI, critical technologies, and secure supply chains, but such cooperation should not require the termination of legitimate economic and technological relations with other partners.

What Is Uzbekistan Most Likely to Do? I would describe Tashkent’s likely approach as “balanced, multi-vector technology diplomacy.”

The country will probably seek to:

1. Maintain Chinese investment and access to Chinese technologies where they support national development goals;

2. Expand cooperation with the United States and Western partners in advanced and strategically important technologies;

3. Diversify suppliers so that no single foreign company controls critical digital infrastructure;

4. Develop national capabilities in AI research, data governance, cybersecurity, and digital infrastructure;

5. Separate commercial technology cooperation from genuinely sensitive national security technologies;

6. Avoid formal commitments that unnecessarily limit future choices of partners;

7. Use competition among major powers to secure better investment conditions and technology transfer.

Therefore, Uzbekistan’s likely response to a U.S. approach will not be a simple “yes” or “no.” The position will more likely be: “We are ready to cooperate with you, but the terms of that cooperation must protect our national interests and preserve our ability to work with other partners.”

This is consistent with Uzbekistan’s broader diplomatic trajectory. Overall Assessment: From AI Blocs to the Struggle for Technological Autonomy. The most important issue here is not simply the emergence of WAICO or the fact that an American letter is being prepared.

The bigger development is the gradual transformation of artificial intelligence into an instrument of geopolitical and economic power. This is also why the issue should be viewed beyond technology alone: as the Global Academy for Future Governance (GAFG) has emphasized in its work on emerging technologies and governance, technological connectivity increasingly depends on the institutional frameworks, regulatory coherence, and international cooperation surrounding it. AI competition is increasingly connecting issues that were previously treated separately: Semiconductors—critical minerals—energy—data centers—
telecommunications—cloud infrastructure—research—education—standards—investment.

For Central Asia, this creates a complex situation. The region lies between major geopolitical and economic centers while remaining interested in cooperation with all of them. It is therefore important to be cautious about describing the future as a simple confrontation between an “American bloc” and a “Chinese bloc.”

A more likely scenario is a period of managed technological competition, in which countries seek to maintain ties with several ecosystems while selectively complying with restrictions in sensitive areas.

Central Asian states are likely to continue defending this flexibility. For Uzbekistan, the key goal should not be technological neutrality for its own sake, but technological diversification and resilience. The country should avoid dependence on a single foreign supplier for critical infrastructure, computing resources, data, or AI systems.

At the same time, it should not automatically exclude technologies simply because they originate from one of the major geopolitical centers. The central strategic principle can be stated very simply:

Cooperate broadly, depend narrowly, and build domestic capabilities wherever possible. If major powers genuinely want to strengthen their position in Central Asia, they should compete not primarily for the political loyalty of regional governments, but for the quality of the technologies, investments, and partnerships they offer.

The task for Central Asia is to achieve the opposite outcome: to ensure that U.S.-China technological competition increases the region’s bargaining power rather than reducing its strategic choices.

Ultimately, the fundamental question for Uzbekistan and its neighbors is not: “Which AI bloc should we join?”

It is: “How can we participate in the global technology economy and the emerging international system for AI governance while preserving our strategic autonomy?”

This will likely become one of the major challenges for Central Asia’s technology diplomacy in the years ahead.