Sovereign AI describes artificial intelligence systems whose models, training data, and compute are owned and governed by the organization, nation, or individual that operates them, rather than rented from a third-party cloud or foreign provider. It emphasizes data control, local or on-device inference, open model weights, and computational independence.
Projects in this space include self-hosted and local-first assistants, on-device inference runtimes, privacy-preserving and federated training, and tooling for regulatory compliance such as the EU AI Act. The shared goal is to reduce external dependencies and keep sensitive data and decision-making under the operator’s own control.