Security
Controls, stated plainly.
Every control on this page is labelled by what is implemented today, what is configured per brokerage, and what is not built. No control appears without a label.
Controls ledger
Nine controls, three labels. Read the chip before you read anything else.
- Transaction-scoped accessImplemented
- Role-based permissionsImplemented
- Recorded review decisions with account and timestampImplemented
- Broker-level controls over connected systemsPer brokerage
- Required-document policyPer brokerage
- Export, retention, and deletion agreed in writingPer brokerage
- Single sign-onNot built
- Email and phone system connectionsNot built
- CRM and transaction-platform integrationsNot built
Implemented: live in the product today. Per brokerage: set with each brokerage during setup. Not built: not part of the product today.
Access levels
Three levels. Each sees only what it should.
- ConsumersSee only what their brokerage shares.
- AgentsSee their own files.
- Brokers and authorized staffSee the company.
Human oversight
AI output is reviewable and attributable. Every review decision carries the account that made it and the time it was made. Nothing unconfirmed drives client communication.
What we do not claim
Terriv holds no third-party security certification and claims none. Architecture and data-handling documentation is available for review during demo and pilot discussions, and questions from IT contacts or E and O carriers are answered directly.
Security questions answered before a demo.
Bring your IT contact. We answer first, then show the product.