Security & trust
Security is a design constraint, not a badge.
Below, what we design for, and separately, what is still on the roadmap. We show no certification seals because we hold none.
Group 1
How we design Lintel
Design principles for a product in early access.
Encryption in transit and at rest
Data is designed to be encrypted on the wire and on disk.
Tenant isolation
Each customer’s data is logically separated. Retrieval is scoped to a project.
Role-based access and project scoping
People see only the projects and actions they are granted.
Audit logging
Sensitive reads and writes are recorded with who, what and when.
Provenance for AI outputs
Every extracted field and answer carries its source page and confidence.
Human approval before external sending
Lintel prepares drafts. Nothing leaves without a person approving it.
No training on customer data
Customer data is never used to train models.
Designed for hosting in India
Designed for AWS Mumbai. Details are confirmed with each customer in early access.
Data export and deletion controls
You can export your data and request its deletion.
Group 2
Roadmap and certifications
| Item | Status |
|---|---|
| SOC 2 | Planned, not yet certified |
| ISO 27001 | Planned, not yet certified |
| Private / VPC deployment | Planned |
| Published sub-processor list | To be published |
AI use
AI-use policy summary
- AI proposes; people decide. Every proposal is reviewable and cites its source.
- Answers are limited to the project’s own documents and decline when unsupported.
- Customer data is not used to train models.
- Outputs are decision support, not legal advice.
Security pack
Request our security questionnaire pack
Start with one contract.
Join early access, or book a walkthrough and we will show you how a single agreement becomes a live register.