Governments and regulated enterprises are increasingly demanding AI systems they can run entirely within their own borders, sparking a global shift away from cloud-based AI services toward sovereign, locally controlled alternatives. Three major developments underscore this trend: a Dubai-based startup launching a platform for governments to run AI on-premises, Russia codifying "sovereign AI" into law with plans to export it globally, and the UK investing £100 million to back homegrown AI companies that serve public sector needs (Sources 1, 2, 3).

What Does "Sovereign AI" Actually Mean?

Sovereign AI refers to artificial intelligence systems that operate entirely within an organization's or nation's own infrastructure, rather than relying on external cloud providers. The concept addresses two core concerns: data security and operational independence. For governments, critical infrastructure operators, and financial institutions, moving sensitive data or workflows to third-party cloud services can violate internal controls or national data sovereignty rules. Sovereign AI lets these organizations retain complete control over where their data lives, which models they use, and how AI systems behave.

Airrived, a Dubai-based company, launched a Sovereign AI Platform designed specifically for this use case. The platform allows customers to build, deploy, and operate AI agents on-premises, on private graphics processing unit (GPU) infrastructure, or in fully air-gapped environments, meaning systems with no connection to external networks. At its core is Airrived's Agentic Operating System, which orchestrates AI agents and applications while keeping the entire intelligence stack under customer control.

"AI sovereignty isn't simply about where your data is stored. It's about who controls the entire intelligence stack," said Anurag Gurtu, Co-founder and Chief Executive Officer of Airrived. "Enterprises should be able to own their data, choose their models, operate their agents and control their infrastructure, without sacrificing the power of agentic AI."

Anurag Gurtu, Co-founder and Chief Executive Officer, Airrived

The platform brings together model management, agent orchestration, enterprise context, reasoning, governance, observability, and AI applications in a single architecture. Customers can use pre-built AI applications or create their own agents while retaining control of the underlying infrastructure. This approach addresses a growing concern as businesses scale AI beyond pilot projects: the cost exposure from consumption-based pricing tied to external model usage. By running workloads within their own perimeter, organizations can achieve more predictable spending and direct oversight of computing resources.

Why Are Governments Pushing for Sovereign AI?

The demand for sovereign AI reflects broader geopolitical tensions and regulatory pressures. In the Middle East, governments have pursued national AI strategies alongside demands for domestic control over digital infrastructure. State-backed initiatives increasingly stress the need to keep strategic data, computing resources, and key systems within national or organizational boundaries. This has created an opening for suppliers offering AI systems that operate without sending data to external cloud environments.

Russia has taken this concept further by embedding it into law. On September 1, 2026, Russia's first federal AI law took effect, establishing a legal framework around what Moscow calls "sovereign AI." The law creates two exclusive categories: "sovereign" and "national" models. Both must generate responses and store data in Russia-owned data centers located in Russia. A sovereign model must stay under the control of a Russian developer throughout its life cycle and be fully reproducible. A national model has more flexibility, allowing the use of foreign open-license components as long as they remain under the Russian developer's control.

Both categories must also pass a government compliance process that checks whether they align with Russian law and what officials describe as "traditional Russian spiritual and moral values." The law does not provide developers with a blacklist of forbidden prompts or dictate exactly how models should answer sensitive questions. Instead, its influence comes earlier in the development process. Developers seeking recognition as sovereign or national must demonstrate that their models comply with Russia's existing speech restrictions and ideological standards.

How Are Different Countries Approaching Sovereign AI?

The UK has taken a different approach, focusing on supporting domestic AI companies rather than imposing strict regulatory controls. The UK Government launched a £100 million competition designed to help British AI companies develop technology that improves public services. The funding is part of the Sovereign AI programme, which was launched in April 2026 and helps promising British AI start-ups establish themselves in the UK, grow their businesses, and compete in international markets.

The first competitions focus on several strategic areas:

• NHS Productivity: AI companies will develop systems that automate administrative workflows, coordinate care, and support decision-making across health services to reduce pressure on healthcare staff and tackle lengthy waiting lists.
• AI Computing Infrastructure: Led by the Department for Business, Innovation, Science and Trade and ARIA's Scaling Inference Lab, this challenge supports technologies that reduce the cost and improve the efficiency of AI computing to increase the UK's public AI computing capacity.
• Defence Applications: Companies will work with the Ministry of Defence to develop technology that securely connects data and advanced AI systems across defence environments to improve operational effectiveness and strengthen national security capabilities.
• AI Security and Resilience: Developed in partnership with the National Cyber Security Centre, this competition supports technologies designed to identify, manage, and reduce security risks as AI systems become increasingly capable of carrying out tasks independently.

The UK scheme has been designed to give smaller AI businesses a better opportunity to compete for public sector contracts. Start-ups often struggle to secure government work because they lack the turnover, financial reserves, or track record expected from larger suppliers. Successful businesses will also keep the intellectual property they create through the projects, meaning they can develop their innovations into commercial products and sell them to customers in Britain and overseas.

What Are the Practical Implications of Sovereign AI?

The shift toward sovereign AI raises important questions about what organizations actually gain and what risks they might face. For enterprises, the appeal is clear: operational control, predictable costs, and compliance with data sovereignty rules. However, the reality is more complex. Russia's approach illustrates the paradox: Moscow plans to export the promise of technological independence using systems that rely on foreign components and a supply chain it cannot fully control. In May 2026, Sberbank, the state-controlled bank leading Russia's AI development effort, said it was looking for Chinese chips to power GigaChat, one of Russia's leading AI models.

A fundamental question remains unanswered: What would a foreign government actually own or control when buying Russian AI? Does a government get copies of the model that it can operate, modify independently, and run on local infrastructure? Or can it only access a Russia-based service? Who manages updates, content filters, audit access, customer data, and decisions about suspending service? Relying on Russian models could give Moscow influence even if there is no hidden malicious capability in their models. Russian providers work under laws that give the state extensive powers of surveillance and access to data.

How to Evaluate Sovereign AI Solutions for Your Organization

• Infrastructure Control: Verify whether the solution allows you to run AI entirely on-premises, in isolated networks, or on private compute infrastructure without external dependencies or cloud reliance.
• Data Governance: Confirm that you can apply access controls and governance rules across AI operations, which matters for internal audit, security teams, and regulatory compliance requirements.
• Model Ownership: Ensure you retain the ability to choose your own models, modify them independently, and maintain full control over the underlying intelligence stack rather than relying on a provider's updates or decisions.
• Transparency and Compliance: Assess whether the provider operates under legal frameworks that protect your data from foreign government access and whether the solution complies with your nation's data sovereignty and regulatory requirements.

The broader question for the sector is whether enterprises will continue to centralize AI around large public cloud model providers or build more localized systems for specific workloads. Airrived argues that for agentic AI, particularly where software is expected to take actions rather than simply answer prompts, some customers will prefer to keep the full stack under their own control.

Russia's AI ecosystem lags significantly behind those of the United States and China. In Stanford's 2024 Global AI Vibrancy Ranking, Russia ranked twenty-eighth out of thirty-six countries when it came to research, talent, investment, infrastructure, and policy. Russian-language AI models still trail leading models. Russia spent about 0.94 percent of its GDP on research and development in 2023, while China spent 2.6 percent. Scientific emigration and cuts to civilian research since Russia's full-scale invasion of Ukraine have further weakened its research base.

Despite these technological limitations, Russia is aggressively marketing sovereign AI abroad. Sberbank's First Deputy CEO Alexander Vedyakhin told Reuters in June that Sberbank was focusing on countries in Africa, Asia, Latin America, and Oceania that want their own AI but cannot afford to build it themselves. Vedyakhin acknowledged a trade-off: the first systems might be slower and less advanced than top foreign models, but they are more affordable, can be adapted locally, and meet state needs.

The sovereign AI movement reflects a fundamental shift in how governments and enterprises think about technology strategy. Rather than accepting dependence on US or Chinese tech giants, organizations worldwide are investing in systems they can control, modify, and operate within their own borders. Whether this approach delivers genuine independence or simply shifts technological dependence to different providers remains an open question, but the trend is unmistakable.