Data protection,
built into the platform.
Security is not a wrapper around an MJDOIT system — it is designed into how claims, data and documents move through it.
Security controls
Identity & Access
Role-based access control and scope-limited visibility. Users see only what their role permits — down to the individual document, team or action.
PII Masking & Data Protection
Sensitive personal information is masked before external AI processing. Sessions use CSRF origin protection and strict token practice.
AI Governance
Whether the model runs in-cloud or on-premise, defined prompt boundaries keep instructions within scope and consistent with your policy.
Document Integrity
Certificates and contracts carry hash-verified serial numbers. Official deliverables carry a tamper-evident receipt.
Who does what — and why it matters.
Separation of duties runs through the whole model: the agent who writes is never the agent who approves. Every release passes an independent review gate, and the audit ledger keeps the permanent record.
The controls behind every claim.
Each security statement on this page maps to a concrete control. The table below is what we review with your security team before onboarding.
| Control | Implementation |
|---|---|
| Identity & Authentication | Signed sessions (HMAC token), role-scoped login, session expiry and IP binding. |
| Authorization (RBAC) | Every action is checked against the user role and object scope — down to document, team or action. |
| Data Flow | Raw documents stay in your boundary; only masked, task-scoped data reaches external models. |
| Encryption | TLS in transit; data-at-rest encryption on storage. Keys are managed per deployment. |
| PII Masking | Sensitive personal information is masked before external AI processing and restored on delivery. |
| Prompt Boundary | Hard policy boundaries keep AI instructions within scope and consistent across sessions. |
| Logging | All key actions are logged and signed into the audit ledger with actor, entity and timestamp. |
| Retention | Configurable retention policy per customer; default aligned with document lifecycle. |
| Deletion | Guaranteed deletion of customer data on request or contract end, including derived outputs. |
| Backup | Encrypted backups with recovery point objective defined per deployment SLA. |
| Model Provider | Cloud, on-premise or local model — the choice is yours; provider is disclosed in the agreement. |
| Data Residency | Deployment region is selected by the customer; on-premise option keeps all processing inside your network. |
| Incident Response | Named response contact, defined escalation path and notification within the agreed window. |
This coverage is confirmed in our security appendix before onboarding and re-verified during the audit review.
Discuss your security requirements.
We share architecture details, deployment modes and data handling practices with prospective partners.