Security & trust

Model governance

Local determines where AI runs. Governance determines whether it is safe, grounded and approved. Models, embeddings, retrieval configurations and prompt policies are all governed components.

Model abstraction

A model-adapter layer keeps the platform model-neutral: open-weight defaults or a customer-provided endpoint. Swapping models is a governed change, not a rebuild.

Evaluation before use

Versioned, checksummed golden sets measure grounding, citation coverage and refusal recall before a configuration is approved for operational use.

Regression after change

Model, embedding or knowledge updates trigger re-evaluation. A configuration that regresses does not activate.

Human oversight

Output is advisory and labelled with confidence and risk indicators. The system refuses when approved knowledge is insufficient and never executes operational actions.

Known limitations, stated plainly

  • Injection resistance is measured in evaluation; no claim of perfect protection or zero hallucinations is made.

These are architecture and control descriptions, not audit results. Current assurance status — implemented, tested, planned — is summarised in the Security & Trust Centre, and deeper material is available through the security architecture pack.

Deploy private AI without surrendering control.

A briefing takes one call. No obligation, NDA available on request, and technical or executive sessions to suit your team.