Aspen Digital

For much of the last decade, the race for AI dominance has been framed as a binary competition between the US and China. However, this narrative overlooks a crucial factor: the role of the Global South in shaping the future of AI. Emerging economies in Africa, Latin America, and Southeast Asia are not just passive recipients of AI technologies; they are actively influencing the direction of AI development, adoption, and governance.

These nations are setting regulatory precedents, providing diverse datasets that enhance AI models, and determining the geopolitical balance of AI power through their technological alliances. Whilst China has expanded its AI footprint in these regions through investments and cost-effective AI solutions, the US and its allies have struggled to offer compelling alternatives. As these nations assert greater control over their AI futures, they are shifting AI geopolitics beyond a simple US-China rivalry, making them not just arenas of competition but key actors in defining the AI landscape.

Why the Global South Matters

The Global South is already playing a key role in the AI landscape. Representing approximately 85% of the world’s population, these regions collectively form a significant portion of the global digital economy and workforce. Despite their growing influence, many nations in the Global South remain ‘tech takers’ reliant on external AI infrastructure and platforms developed by major AI powers like the US and China. They are no longer just passive adopters: their growing populations, increasing digital infrastructure, and expanding economies have helped their policymakers become more proactive in shaping AI governance.

EMERGING A.I. POWERHOUSES

Nations in Africa, Latin America and Southeast Asia are setting regulatory precedents, developing homegrown AI capabilities, and influencing global AI norms in multilateral institutions. For instance, Brazil’s AI strategy emphasises ethical AI governance and innovation in agriculture, while Indonesia’s AI roadmap focuses on workforce transformation and smart cities. Meanwhile, Kenya has positioned itself as a leader in AI-driven financial inclusion, leveraging machine learning for mobile banking and digital payments. Beyond policy, these regions offer AI models unique training environments that enhance adaptability and efficiency. AI-driven medical diagnostics, such as Google’s tuberculosis detection AI trained on African medical datasets, have demonstrated improved accuracy in local healthcare settings. In Latin America, AI-powered climate prediction and crop disease detection models are being refined using agriculture data tailored to smallholder farmers’ needs. Meanwhile, natural language processing models training on Southeast Asian languages, such as those from the SEA-LION project, are improving AI-driven translation and speech recognition for multilingual populations.

These nations also hold considerable sway in multilateral governance institutions, where they represent the global majority and play a critical role in shaping international AI norms and regulatory frameworks. The United Nations, through initiatives such as UNESCO’s AI Ethics guidelines and the UN AI Advisory body, and the OECD AI Policy Observatory, which includes non-member countries, serve as key platforms for Global South engagement in AI governance. Additionally, regional efforts such as the African Union AI Working Group and ASEAN’s AI Governance and Ethics Framework further illustrate how these nations influence AI policy. As AI becomes deeply integrated into governance, security and economic planning, how these countries choose to align, whether with China, the US, or a diversified approach, will heavily influence the future structure of global AI power.

China’s proactive A.I. Outreach with the Global South: Strategic Expansion Through Accessibility

A defining characteristic of China’s AI diplomacy is its courtship of the Global South, leveraging the affordability and scalability of its technology to deepen strategic ties. Companies like DeepSeek, Baidu, and Huawei have actively positioned themselves as affordable partners for nations seeking to modernise governance, infrastructure, and industry with AI-driven systems. DeepSeek, in particular, has emerged as a formidable player, offering an alternative AI ecosystem that prioritises accessibility and integration into state-led digital transformation strategies. Unlike US-led AI systems, which often come with high licencing fees and require extensive infrastructure investments, Chinese AI systems are designed for efficiency and ease of adoption.

A.I. Ecosystems: Adoption vs Indigenous Development

At the core of any AI ecosystem are several interdependent components: data collection and governance, model development, AI infrastructure, deployment, and integration into industries and governance. China’s AI engagement in the Global South focuses primarily on the latter stages, model deployment, infrastructure provision and AI-powered services, rather than fostering indigenous AI development in host nations. Companies like DeepSeek, Baidu and Huawei offer turnkey AI solutions that enable governments and businesses to rapidly adopt AI applications without requiring deep domestic expertise. These solutions include smart city technologies, AI-powered surveillance systems, language processing models and automation for public services.

The Digital Silk Road: A.I. Dependence and China’s Strategic Influence

Through the Digital Silk Road, China has embedded AI capabilities within critical infrastructure projects, fostering long-term dependency on its technology. AI-powered surveillance systems, smart city technology, and digital infrastructure are becoming integral to governance structures across Africa, Southeast Asia, and the Middle East. While China does provide some knowledge transfer, such as AI research hubs and training programs, its emphasis is on embedding Chinese AI models and infrastructure within local ecosystems rather than enabling fully independent AI development.

For example, under the Digital Silk Road, China supplies AI-driven cloud computing, data centres and software solutions that create long-term dependencies on Chinese providers. This approach ensures that while host countries gain access to cutting-edge AI capabilities, they remain reliant on Chinese-developed models and platforms, rather than developing fully autonomous AI industries. Unlike initiatives focused on building domestic AI talent pipelines and model development from the ground up, China’s strategy is centred on scaling AI adoption through direct access to its ecosystem, ensuring continued influence over digital and governance frameworks in partner nations. This level of digital entanglement ensures that China’s influence extends beyond economic ties, embedding itself in the operational frameworks of these nations.

However, this growing dependence raises significant concerns for both the Global South and the West. First, many of the AI-powered solutions China provides, grant Chinese firms access to vast amounts of sensitive government, business and personal data in host countries. This raises concerns over data exfiltration, mass surveillance, and cyber vulnerabilities, particularly as it is at times hard to distinguish between Chinese firms objectives and those of Beijing.

Second, by embedding AI into governance and infrastructure, China is shaping policy, digital governance, and regulatory norms in host nations. This creates long-term dependencies, limiting the ability of those nations to develop independent AI ecosystems and potentially coercing them into aligning with China’s political and economic priorities. Finally, for Western nations, the expansion of Chinese AI into the Global South erodes democratic digital governance models, enabling authoritarian AI practices, such as surveillance-driven governance and content control, to spread. It also limits Western engagement in setting global AI norms, giving China greater influence over how AI is regulated, deployed and secured internationally.

China’s AI investments offer it strategic diplomatic advantages. Despite the associated risks, many recipient countries view Chinese AI solutions as a means to leapfrog traditional development barriers. These partnerships could reshape regional power structures, positioning China as a crucial technology partner in Global South governments’ long-term strategic planning. Chinese companies also actively engage in knowledge transfer initiatives, establishing AI research hubs and training programs in host countries. This not only expands China’s technology footprint but also cultivates goodwill and reliance on Chinese AI expertise.

Trump’s Second Term: The U.S. Retreats from Global Tech Engagement

At the same time, the US is disengaging from tech diplomacy with the Global South. This marks a stark shift from the Biden administration’s approach, which prioritised multilateral AI cooperation and sought to balance AI innovation with responsible governance. While Biden’s policies encouraged engagement with global partners, the Trump administration has pivoted toward a more unilateral, competition-driven strategy.

Since returning to office in 2025, the Trump administration has pursued an aggressive AI innovation policy focused on reducing regulatory constraints and accelerating AI development to maintain US technological dominance. This approach prioritises economic nationalism and private sector-led expansion, aiming to outpace China through deregulation and infrastructure investment. Trump’s Executive Order ‘Removing Barriers to American Leadership in Artificial Intelligence’ mandates the creation of an AI action plan within 180 days. While the administration is loosening domestic regulatory constraints to fuel AI development, it is simultaneously tightening export controls on advanced AI technologies, specifically to limit China’s access. This dual approach reflects a broader strategy: maintaining US technological dominance by restricting adversarial access while accelerating domestic innovation.

As part of this strategy, the administration launched ‘Stargate’, a partnership with OpenAI, Oracle and SoftBank, investing up to $500 billion in domestic AI infrastructure by 2029. This underscores a shift toward infrastructure-driven AI supremacy, accelerating a broader trend in US policy.

The Case for Open-Source A.I. Models

Stargate’s focus on proprietary AI models and national security priorities raises concerns about its competitiveness in the Global South. Unlike DeepSeek which offers scalable AI solutions with low entry barriers, the US remains centred on high-end, privately-driven AI development. To effectively counter China’s influence, Washington may need to incorporate open-source AI development into Stargate, allowing emerging economies greater flexibility in AI adoption while reducing dependence on Chinese infrastructure.

China’s AI strategy has evolved, with companies like DeepSeek embracing open-source models that make elements of their technology widely accessible. DeepSeek’s open-source AI models facilitate rapid adoption across diverse industries, enabling developers in the Global South to tailor AI solutions to their needs. This open approach challenges traditional perceptions of China’s AI strategy as purely proprietary and state contained.

Open-Source A.I. as a Strategy: A U.S. Response to China’s Global AI Expansion

If Washington remains committed to a light-touch regulatory approach to AI in the name of fostering innovation, it must also recognise the strategic imperative of open-source AI. Embracing open-source models would require a shift in policy, including government-backed AI research initiatives, public-private partnerships and revised export controls that balance openness with security. A strategic open-source AI policy could enhance transparency, accelerate technological progress, and provide an alternative to China’s AI expansion in the Global South. This would involve federally funded open-source AI programs, support for regional AI governance initiatives and technical assistance to help emerging economies build independent AI ecosystems.

However, such an approach is not without risks, greater openness introduces vulnerabilities to adversarial exploitation, and raises complex challenges in ensuring responsible deployment. Major US tech firms such as Meta and Google have initiated open-source AI projects, but a more coordinated government-backed effort is needed. By positioning itself as a leader in ethical, adaptable, democratised and safe AI development, the US can counter China’s influence while promoting an AI ecosystem aligned with democratic values and transparency.

Supporting open-source AI would also enable emerging economies to develop independent AI ecosystems, reducing their reliance on foreign technology providers – a shift that may not necessarily align with immediate US commercial interests, but would offer clear advantages to developing nations, while reinforcing democratic AI governance.

Rethinking A.I. Strategy: U.S. Investment Without Traditional Foreign Aid

However, with the Trump administration’s drastic cuts to the USAID budget and programming, alternative mechanisms would be necessary to implement this strategy. Rather than relying on traditional development assistance, Washington could expand strategic investments in AI research partnerships, education programs, and cloud-based open AI platforms through agencies more aligned with Trump’s priorities. 

The National Science Foundation (NSF) and DARPA could take the lead in funding open-source AI research, while the Department of Commerce, through its National Institute of Standards and Technology (NIST) could establish interoperability standards to ensure that open-source AI aligns with security and ethical best practices. Given the administration’s emphasis on economic nationalism, the International Development Finance Corporation (DFC) and the Export-Import Bank (EXIM) could be leveraged to support AI infrastructure investments in the Global South.

Additionally, federal investment in cloud-based AI platforms, modeled after domestic AI infrastructure initiatives like Stargate, could offer emerging economies secure, scalable alternatives to Chinese AI solutions, without relying on traditional foreign aid mechanisms.

This approach would not only foster a more competitive and decentralised AI landscape but also ensure that AI development in the Global South aligns with open, transparent, and democratic governance principles, all while sidestepping commitments to multilateral engagement. 

The Challenge of Multilateral A.I. Governance

As AI capabilities evolve, governance structures must keep pace. The Global South, which represents the majority in multilateral forums, is no longer a passive participant but a decisive force in shaping global AI norms. With priorities ranging from digital sovereignty to equitable AI access, these nations are asserting their influence in governance debates, driving independent initiatives, and challenging traditional power hierarchies.

A clear example of this was Brazil’s leadership in UNESCO’s AI Ethics Guidelines, where it played a pivotal role in advocating for principles of data sovereignty and equitable AI access, ensuring that AI policies consider the needs of developing economies. Yet, while China actively courts the Global South with scalable AI solutions and governance partnerships, the US has deprioritised multilateral AI diplomacy. The decision to halt USAID funding and abstain from international AI agreements, evidenced by its decision, along with the UK, not to sign the Paris AI summit statement, reflects a broader shift towards prioritising domestic AI development over global engagement.

Vice President JD Vance’s emphasis on minimal regulation and American leadership signals a belief that US innovation alone will maintain AI supremacy. However, this fragmented approach weakens Western cohesion, creating space for China to shape AI governance frameworks that reinforce its technological and political influence.

For many emerging economies, AI adoption is not merely a technology shift, but a matter of digital sovereignty. Many governments increasingly recognise the risks of over-reliance on foreign AI providers, including data security vulnerabilities and economic dependencies that could limit their strategic autonomy. In response, nations such as India and Brazil are investing in domestic AI capabilities, while regional bodies like the African Union are crafting governance frameworks that align AI adoption with local ethical and regulatory principles.

Strategic A.I. Autonomy: A U.S. Path to Competing Without Forcing Dependence

Rather than retreating from multilateral governance, the US should champion strategic AI autonomy, not full detachment from global AI ecosystems, but ensuring that nations have access to trusted and diversified AI models that do not force them into dependence on China’s AI infrastructure. This does not mean undermining US AI firms, but rather shaping AI partnerships that provide market access while offering a compelling alternative to China’s state-backed AI exports. A balanced strategy would focus on capacity-building partnerships, secure cloud-based AI platforms and interoperable AI ecosystems that allow emerging economies to develop their own AI capabilities while still benefiting from US-led innovation and infrastructure.

Some nations are also leveraging AI to strengthen regional cooperation. ASEAN, for instance, is developing a regional AI framework to establish standardised governance policies across member states. This initiative aims to balance AI adoption with regulatory oversight, preventing monopolistic control by any single external actor, whether China or the US. Similarly, Latin American nations are collaborating on AI ethics guidelines tailored to their socio-economic contexts, ensuring that AI development aligns with national interests, rather than foreign corporate or geopolitical imperatives.

By sidelining multilateral AI governance, Washington risks not only losing influence to Beijing but also neglecting the growing demand from the Global South for AI models and policies that reflect their economic realities and governance priorities. Given that developing nations represent the majority in multilateral discussions, their choices will shape the future of AI governance. If the US fails to engage, emerging economies will look elsewhere, shaping an AI trajectory that does not align with US and its allies’ strategic interests.

A.I. Leadership: Will the U.S. Cede the Future to China?

The future of AI geopolitics will not be dictated solely by the US or China, but by the choices of the emerging AI economies of the Global South. However, as Washington retreats from multilateral AI diplomacy and fails to offer compelling alternatives, China is rapidly filling the vacuum. Through strategic investments, scalable AI solutions and deepening technological dependencies, Beijing is embedding its influence in the digital infrastructure and governance frameworks of the Global South.

If the US does not act decisively, by investing in open-source AI, capacity building and a more proactive AI foreign policy, it risks not just losing influence, but allowing China to set the rules of global AI governance unchallenged. Without a strategic response, Washington is not just stepping back from AI leadership, it is actively enabling China to define the future of global AI governance and technology provision on its own terms.

The views represented herein are those of the author(s) and do not necessarily reflect the views of the Aspen Institute, its programs, staff, volunteers, participants, or trustees.