Governments and companies worldwide are racing to build their own artificial intelligence systems after the Trump administration's restrictions on Anthropic demonstrated how geopolitical tensions can disrupt access to critical AI tools. The move has transformed what was once a niche concern about technological independence into an urgent priority for national security and economic resilience across multiple continents.

What Exactly Is Sovereign AI and Why Does It Matter Now?

Sovereign AI refers to artificial intelligence systems developed and operated domestically to reflect a country's language, data, laws, and security requirements. Unlike relying on platforms controlled by foreign companies or governments, sovereign AI keeps critical data and decision-making power within national borders. The concept gained sudden urgency in June 2026 when the U.S. Commerce Department ordered Anthropic to suspend access to its AI tools for all foreign nationals, including employees working for the company. Though the restrictions were lifted less than three weeks later, the message was clear: the U.S. government is willing to use trade restrictions to control who can access the most powerful AI systems.

The parallels to historical trade routes are striking. As one technology analysis notes, the situation mirrors the Age of Exploration, when emerging maritime powers realized they could not depend on established trade networks controlled by others and instead invested in forging their own routes. Today's companies face the same fundamental question: should they entrust their most critical assets, including customer data and proprietary technologies, to massive external platforms without fully recognizing the risks?

How Are Countries Building Independent AI Infrastructure?

• South Korea's Defense Partnership: Naver, one of South Korea's leading AI companies, partnered with Korea Aerospace Industries to develop sovereign AI optimized for the country's defense and security environment. The collaboration will focus on creating a defense-specific AI foundation model and applying AI to unmanned aircraft platforms and AI pilot development for future combat systems.
• Private Sector Leadership: Companies are moving beyond passive users of massive platforms to become autonomous sovereigns capable of dynamically creating value. An automotive manufacturer, for example, can securely connect AI systems across global production sites and suppliers, allowing alternative manufacturing routes to be identified immediately if geopolitical risks disrupt component supplies.
• Closed-Environment Training: Pharmaceutical companies can now train AI on highly confidential genomic and clinical trial data within fully closed environments, enabling rapid identification of promising drug candidates without exposing sensitive information to external platforms.

South Korea's initiative is particularly significant because it represents a direct response to the Trump administration's actions. Naver CEO Choi Soo-yeon stated that "technological self-reliance in national defense and security is directly connected to national sovereignty, making it essential to secure independent sovereign AI infrastructure". The partnership aims to strengthen South Korea's defense technology sovereignty and create new global competitiveness for the future defense industry.

Could These Sovereign Systems End Up Using Chinese Technology?

Perhaps the most surprising consequence of U.S. restrictions is that countries seeking AI independence may turn to Chinese models rather than building entirely from scratch. A 2026 AI Index report from Stanford University's Institute for Human Centered AI found that the performance gap between U.S. and Chinese AI models has effectively closed. China's DeepSeek briefly matched the top U.S. models on a key benchmark in 2025, and the best model from Anthropic is now only a few percentage points ahead of strong Chinese models.

Many Chinese AI models are "open source," meaning companies, universities, and governments can download and run them on their own hardware rather than relying on servers hosted in another country. This creates a paradoxical outcome: the Trump administration's attempt to restrict access to U.S. AI technology may inadvertently accelerate the adoption of Chinese AI systems globally. If the future is AI sovereignty, it may well be built on Chinese code rather than U.S. systems.

The broader implication extends beyond AI alone. A report by the Canadian Anti-Monopoly Project found that 85 percent of Canada's cloud computing resources are controlled by three U.S.-based companies: Amazon, Microsoft, and Alphabet (Google's parent company). The number is close to 70 percent in most nations. The Trump administration's conflict with Anthropic may trigger not just a wave of national AI projects, but also broader national information technology infrastructure development.

What Does Sovereign AI Actually Look Like in Practice?

Fujitsu, a company with decades of experience building national-level infrastructure systems for governments and financial services, has outlined a practical approach to sovereignty. The company's Takane is a domain-specific large language model (LLM), which is a type of AI trained on specialized information, that operates in secure, private environments isolated from external systems. Customer-critical data and proprietary knowledge remain fully protected, while organizations retain flexibility to deploy AI according to their own operational rules without being constrained by external platform limitations.

A complementary technology called Multi-AI Agent Technology enables multiple AI agents to collaborate securely while maintaining their individual autonomy. This architectural approach optimizes outcomes at a system level and serves as a core technology within the envisioned Sovereign AI Platform for real-world deployment. The result is that companies move beyond the role of passive users of massive platforms and evolve into autonomous sovereigns capable of designing their own futures.

"Technological self-reliance in national defense and security is directly connected to national sovereignty, making it essential to secure independent sovereign AI infrastructure," said Choi Soo-yeon, CEO of Naver.

Choi Soo-yeon, CEO, Naver

The shift toward sovereignty represents a fundamental redefining of competitive rules. Companies that control their own AI systems gain durable competitive advantage, even in an era defined by uncertainty and geopolitical volatility. The pharmaceutical industry example illustrates this clearly: by training AI on confidential genomic and clinical trial data within closed environments, companies can rapidly identify promising drug candidates and open new pathways to lifesaving treatments without exposing proprietary information to foreign entities or platforms.

What Happens to Global AI Development Now?

The Trump administration's approach to AI regulation has inadvertently made the case for AI sovereignty to every technology adviser urging governments from the Republic of Ireland to Israel to develop their own AI tools. Rather than creating a unified global AI ecosystem, U.S. restrictions are fragmenting the landscape into competing regional systems, each optimized for local values, languages, and security requirements.

This fragmentation could ultimately reshape how AI systems are built. As countries like Canada and France develop sovereign AIs, we may see the rise of systems built around carefully chosen national texts and archives, embodying national values and priorities rather than the haphazard collection of internet-scraped data that powers most current models. The result could be a more diverse AI landscape, but one that is also more fragmented and potentially less interoperable across borders.

For companies and governments watching these developments, the message is clear: the era of depending on a single dominant AI platform controlled by a foreign entity is ending. Whether through partnerships like South Korea's defense AI initiative, through proprietary systems like Fujitsu's Takane, or through adoption of open-source Chinese models, the world is moving toward a future where AI sovereignty is not a luxury but a fundamental requirement for maintaining control over critical data and strategic decision-making.