Government

Policy-constrained AI for public-sector programs where oversight, records obligations, and public trust depend on auditable evidence.

Deployment considerations for Government — which tiers fit and how to get started.

Tier 1

Workshop

Governed public-model overlay. Assists judgment without replacing it.

Tier 2

Refinery

Closed inference with sovereign models. Fails closed on authority gaps.

Tier 3

Clean Room

Air-gapped sovereign deployment. Fails hard — no external dependencies.