- Sovereign AI is AI that an organization owns and controls: its agents, models, data and results, run in any cloud, on premises or air-gapped.
- The phrase entered policy in a Tony Blair Institute report in February 2023, and NVIDIA CEO Jensen Huang made it widely known from November 2023.
- There is no agreed definition of sovereign AI: Brookings and CEPS describe AI sovereignty as a spectrum of strategies rather than full self-sufficiency.
- Gartner's 2026 definition of AI sovereignty covers a nation or an organization, so the term now applies to companies as well as countries.
- Data residency covers where data is stored and an air gap covers network isolation; sovereign AI covers who owns and controls the AI itself.
What is sovereign AI?
Sovereign AI means the organization that uses AI also owns it. Its agents, its models, the data they learn from and the results they produce belong to that organization. It also decides where they run: a public cloud, its own data center, or a network with no internet connection. Nothing depends on a provider keeping an API available, priced the same, or unchanged.
The term started with nations and now covers organizations too. Gartner defines AI sovereignty as "the ability of a nation or organization to independently control how AI is developed, deployed, and used." The idea is the same at both sizes: whoever owns the AI controls it.
Where the term comes from
Sovereign AI grew out of digital sovereignty, the idea that a state should control the digital technology it depends on. CNAS calls it "an offshoot of digital sovereignty."
The phrase entered policy in February 2023. A Tony Blair Institute report on Britain's technology strategy warned of "the erosion of the UK's sovereign AI capabilities" and called for government-led "sovereign general-purpose AI systems." A study of 485 news articles by Heesoo Jang dates the first press use to 22 February 2023 and traces its spread to India, South Korea, Germany, Australia and France.
NVIDIA brought the term into wide use. A 2025 review in Natural Language Processing credits CEO Jensen Huang, who "elevated the visibility of the concept, starting with an Nvidia earnings call in November 2023."
There is still no agreed definition. CNAS says a consensus definition "remains elusive." IISS lists at least seven components, from compute and chips to foundation models, talent, and regulation and governance. Researchers treat it as a matter of degree: Brookings and CEPS define AI sovereignty as "a spectrum of strategies" for making independent decisions about AI, rather than full self-sufficiency.
For companies, Stanford HAI finds the term usually means operational control: running AI on premises and avoiding vendor dependencies. That settles where AI runs, not who owns the models or who can change them.
One older meaning is unrelated. In AI-safety writing, a "sovereign" is an AI that acts on its own without human steering, a category from Nick Bostrom's typology of advanced agents. That usage has nothing to do with who owns or controls AI.
How sovereign AI works
Sovereign AI comes down to four things an organization owns, plus the freedom to run them anywhere.
| What you own | What it means in practice |
|---|---|
| Agents | The workflows, tools and permissions your AI agents use are yours to inspect, change and move. |
| Models | You run open or fine-tuned models with weights you hold. A provider can't retire or reprice them under you. |
| Data | Training data, prompts and documents stay inside your boundary. Nothing goes to a third party to be processed. |
| Results | Outputs, logs and audit trails stay with you, so you can show a regulator how a decision was made. |
Where it runs is a separate choice. The same stack can run in a cloud account you control, in your own data center, or fully air-gapped. Air-gapped is one end of that range, not the definition.
Ownership also changes how you pay. A rented large language model is billed per token. An owned model, often a small language model fine-tuned for one job, costs the compute it runs on.
Sovereign AI vs data residency vs air-gapped AI
These three get treated as one idea. They overlap, but each answers a different question.
| The question it answers | What it covers | |
|---|---|---|
| Data residency | Where is the data stored? | The data's location only. A model hosted somewhere else can still process it. |
| Air-gapped AI | Is the system cut off from the internet? | Network isolation. An air-gapped system can still run a model you don't own and can't change. |
| Sovereign AI | Who owns and controls the AI? | Agents, models, data and results, wherever they run. |
Data residency and an air gap can both be part of sovereign AI. Neither is enough on its own. As McKinsey's Ali Ustun puts it, "you can actually have data sovereignty, but you may not have sovereign AI." Residency promises also have legal limits: Lawfare notes that US providers' commitments to keep data in Europe "cannot stand in the way of their legal obligations" under the US CLOUD Act.
Sovereign AI examples
- A federal research agency runs an AI assistant with no public-internet connection. At NIH, it assists 8,000 employees across 28 institutes and deflects up to 10,000 IT service requests a year.
- A telecom operator replaces rented models with its own. AT&T cut costs by up to 90% with fine-tuned small language models.
- A bank keeps customer data and model decisions inside its own environment, so its model risk team can audit every result.
- A national government funds its own compute and models. India approved its IndiaAI Mission in March 2024, with public compute of 10,000 or more GPUs and investment in indigenous foundation models.
Sovereign AI in H2O AI
The H2O Sovereign AI Platform is built on this idea. It runs predictive, generative and agentic AI in one system, and you own the agents, models, data and results. The platform runs in any cloud, on premises or air-gapped. It is open source at its base and runs any model, so there is no lock-in.
Deployment options include H2O AI Managed Cloud, H2O AI Hybrid Cloud and H2O AI Cloud for Government, which is FedRAMP High Certified.
For how H2O puts sovereign AI into practice, see Sovereign AI at H2O.