Your data never leaves
Sensitive paths run on your own models and infrastructure — on-prem or your private cloud.
Security
AI that lives in your perimeter.
Wide Moat runs inside your infrastructure or your private cloud. Every request passes through one governed gateway, access follows your identity provider, and every prompt, cost and action is logged.
Data boundaries
What stays inside, and how.
One governed gateway
Every request is routed, budgeted, rate-limited and logged. Nothing talks to a model on its own.
Access-aware by design
SSO and directory groups flow into every call — people see in AI only what they already can.
Fully observable
Every prompt, cost and action is measured — so adoption and risk are both visible.
Two deployment models
Two deployment models, the same controls.
Governance
One identity, one policy, one audit trail.
Identity
The assistant, agents and internal apps inherit the same user and source-system permissions.
Models and policy
Routing, data boundaries, approvals, budgets and audit apply across the platform.
Human control
Actions can require approval, and exceptions are escalated to a person with the source and reason.
Responsible disclosure
Found a vulnerability?
Write to contact@widemoat.ai. Our disclosure contact is also published in security.txt (RFC 9116).
Security review
Documentation for your questionnaire.
Architecture, data flows, deployment runbooks and the answers to your security questionnaire are available on request. We state what is in place and what is not, with no certifications claimed that we do not hold.
Start with one valuable function
Bring your security team to the first call.
We will walk through the perimeter, the gateway and the access model on your own deployment option.