A government customer requires that no foreign entity can access their data or compel its disclosure. What are the options and what does each cost?

What is being asked

Not only that data is stored in a country, but that no entity subject to a foreign jurisdiction can be compelled to hand it over — which is a question about the operator and the keys, not about the storage location. A provider headquartered elsewhere may be subject to foreign legal process regardless of where the servers sit.

The options

• Sovereign cloud regions operated by a local entity, with local staff and local ownership. Strongest, and it means fewer services, slower feature availability, and a smaller operational scale.
• Customer-held encryption keys, so the provider technically cannot produce plaintext. Practically strong, and it constrains the services that can process the data — most managed services need to read it — and it puts key availability on the customer's critical path.
• Confidential computing, keeping data encrypted in use, which narrows the gap and depends on hardware guarantees and a trust chain the customer must be willing to accept.
• On-premises or private cloud, which is the fallback when nothing else satisfies and reintroduces every operational burden the organisation moved to cloud to shed.

The trade to state honestly

Sovereignty and capability are in tension. The strongest sovereignty options offer the fewest managed services, the slowest access to new capability, and the highest operational cost. This is not a temporary market gap; it follows from the fact that global scale and local exclusivity are opposed.

What frequently resolves it in practice

Tiering by data classification. A minority of the data genuinely requires the strongest treatment; the rest does not. Applying the strictest regime to everything is what makes sovereignty programmes cost several times what they need to — and the classification work that avoids it is cheaper than the infrastructure it saves.