The Rise of Sovereign AI as a Foundation for Government and Enterprise

Sovereign AI is quickly moving from concept to implementation as governments and enterprises seek more authority over data, models, infrastructure and policy. Trusted AI at scale depends on strong governance, resilient infrastructure and the right partnerships — not just access to models.

Across government and enterprise, AI is moving beyond experimentation and into sensitive, high-stakes operations. As that shift accelerates, a new question is rising to the top of the agenda: who controls the data, models, infrastructure and policies behind AI?

That question sits at the heart of sovereign AI. At its core, sovereign AI is the strategic capacity of a government or private enterprise to independently govern, develop and operate its AI lifecycle with full authority over data, compute, models and policy.

Two recent studies, one focused on government leaders and one on enterprise IT decision-makers, point to the same conclusion: sovereign AI is moving from long-term ambition to near-term execution. The drivers may differ by sector, but the pattern is consistent. Organizations want AI that is powerful, scalable, compliant and trusted.

In government, that can mean ensuring citizen data, model outputs and AI-enabled workflows remain governed within national boundaries and approved policy frameworks. In enterprise, it often means protecting proprietary IP, customer data and strategic processes inside an operating environment the organization can control. In both cases, sovereign AI is becoming less about theory and more about operational readiness.

Government adoption accelerates

A global IDC study commissioned by Dell Technologies¹ found that governments are no longer debating AI’s value. They are focused on how to deploy it responsibly and at scale.

• • 71% of government leaders believe agentic AI, systems capable of autonomously completing complex tasks, will accelerate AI adoption in government.
  • 52% plan to invest in Sovereign AI within 12–18 months, signaling a shift from pilot to production.
  • 66% report that technology is evolving faster than their workforce can keep up, creating urgent demand for AI-enabled solutions.
  • 58% identify strong sovereign data governance, quality and control as among the most critical platform requirements for sovereign AI.

Government leaders are also setting clear conditions for how adoption should proceed. Forty-four percent say they will accelerate AI only if strong safeguards around data security, privacy and sovereignty are firmly in place.

Enterprise investment intensifies

The enterprise story mirrors and, in some cases, accelerates what we’re seeing in government. A new study conducted by Omdia surveyed IT professionals across North America and Europe, spanning large midmarket and enterprise organizations across industries.²

The findings show that enterprise leaders already treat sovereign AI as an active investment priority. This is driven by the need to protect data, meet regulatory expectations and scale AI with greater control.

• • 73% of respondents are actively implementing or piloting sovereign AI capabilities. This indicates most responses reflect real-world experience rather than speculation.
  • Nearly one-third (32%) of respondents rank sovereign AI as their highest strategic technology priority, ahead of digital transformation, cybersecurity, generative AI and agentic AI.
  • 41% are allocating $1 million or more to sovereign AI over the next 12 months, with spending expected to increase to an average of $3 million annually.
  • The most important business drivers focus on protecting and controlling proprietary data and IP. Competitive advantage is seen as a longer-term benefit that follows once the data foundation is secure.
  • No single governance strategy dominates. Enterprises are combining policy frameworks, technical controls and operational processes because sovereign AI compliance is complex and cross-functional.

Omdia’s research suggests that sovereign AI is an organization-wide challenge, not a point solution. It requires coordinated investment across infrastructure, governance, operations and monitoring.

Shared priorities are emerging

Despite their different missions, government and enterprise leaders are converging on a common set of priorities.

The first is data control. Across both sectors, organizations understand that AI confidence starts with data confidence. If they do not know where data resides, how it is classified, or who can access it, they cannot responsibly deploy AI at scale.

The second is governance. For government leaders, safeguards must be in place before adoption can accelerate. For enterprises, governance is multi-layered, requiring policy frameworks, technical controls and operational processes that work together. In both cases, trust has to be designed into AI from the beginning.

The third is trust. No one can do this alone. Governments see public-private partnerships as essential to accessing the expertise required for secure AI implementation, but trust is paramount to successful outcomes. Enterprises view partner selection as a systems-level decision where breadth, trust and contextual fit matter more than any single capability. Sovereign AI success depends on trusting the ecosystem, not any individual vendor.

The fourth is workforce readiness. In both sectors, the skills gap is real. Technology is advancing faster than many teams can absorb it, which increases the appeal of supported, turnkey approaches that reduce operational burden without giving up control.

Urgency Is Building

What has changed is not just regulation. It is the scale and persistence of AI consumption.

As organizations move from copilots to autonomous agents, AI starts to behave less like a bounded tool and more like a continuously operating system. In government, agentic AI can act as a workforce multiplier, helping teams automate complex tasks and redirect staff toward higher-value work. In enterprise, always-on autonomous agents can turn sovereign AI from a compliance decision into a cost, control and operating model decision.

That makes predictability more important. As AI usage scales, token-based consumption models can introduce variability in both usage and spend, making costs harder to forecast and govern over time. Sovereign AI offers a path to regain control by giving organizations greater authority over utilization, performance and policy, while turning AI from an open-ended expense into governed infrastructure.

The path forward

Sovereign AI is no longer only about compliance. It is becoming the operating model for trusted AI in regulated and data-sensitive environments.

The research shows that governments and enterprises are converging on the same operational requirements for AI at scale: stronger data foundations, clearer governance, resilient infrastructure and trusted partnerships. As organizations move from experimentation to execution, the challenge is no longer simply access to models. It is creating an AI environment that delivers control, compliance, performance and more predictable economics.

Dell Technologies is focused on helping organizations turn those requirements into a practical path forward, starting with trust. For more than 40 years, customers around the world and across industries have relied on Dell to deliver secure, reliable infrastructure at scale. That experience, combined with Dell’s technology portfolio and partner ecosystem, helps organizations move from strategy to implementation with greater confidence.

^*Disclosure: Findings reflect survey responses from participating organizations and may vary based on region, industry, and implementation maturity.

¹Source: IDC InfoBrief, “Building a Sovereign AI Foundation for Government,” February 2026. Global survey of 258 government IT decision-makers. (sponsored by Dell Technologies in partnership with NVIDIA), #US54253026-IB, IDC Resource Map Document, Building a Sovereign AI Foundation for Government, #US54298426, March 2026

²Omdia Research Survey, The Great Repatriation: Why Enterprises Must Act Now to Ensure AI Sovereignty, February 2026. Global survey of 450 IT decision-makers.