Secure legal operating system with local inference —
fully under your control.
Matter-level isolation, practitioner-controlled approval gates, and a signed, append-only record of every interaction.
Development status
Bernio is in active development with design partners — firms running real matters through the platform and shaping what is built next — and the first commercial contracts are being signed on founding terms. The capabilities described on this page are built and can be demonstrated on a matter type your team handles. Bernio Pro — the on-premise appliance — is an engineering prototype and is not yet commercially available, so any capability on this site that depends on it is marked as such. Descriptions of the product are not warranties; the binding terms for any deployment are those set out in the applicable written agreement.
Every part of a matter has a place to live.
Bernio is one system, but it is not one screen. Each module below owns a stage of the work — and each enforces the same isolation and the same audit.
Three layers. One coherent platform.
Security, workflow logic and AI capability each operate in their own layer — and can be audited, upgraded or replaced independently.
Workflow Engine
Intake, draft, review, approval, escalation, close — modelled as states. AI is invoked at each stage, never in place of it.
Security & Isolation
Every client and matter in its own environment, with access enforced by role and matter context.
Model Orchestration
Task routing, not prompt engineering. Each task is dispatched to the architecture best suited to it.
Legal workflows have structure. The platform should respect it.
A matter has phases, dependencies, required approvals and evidentiary obligations. Bernio models them, so AI operates within the workflow rather than around it.
Practitioner approval at every decision point
AI output is held pending review. Nothing reaches a client, a counterparty or the matter record without a named sign-off.
Structured matter strategy before the first billable hour
Scope, dependencies and jurisdictional risk are agreed with the lead practitioner before execution begins.
Configurable to your firm's standards and precedents
Built once per firm and applied consistently across every matter, practitioner and agent.
The right model for every task. Not one model for everything.
A paralegal-level extraction does not need the model a cross-border risk analysis needs. Bernio routes each task to the architecture optimised for it, within your firm’s approved model policy.
- Claude, ChatGPT, Gemini, Mistral and locally-hosted open models — configured per firm and task type
- Approve, restrict or block models by practice area, matter sensitivity or client mandate
- Every inference records which model version produced which output, on which matter
The on-premise AI appliance for firms that keep everything in-house.
Bernio Pro runs the full platform inside your infrastructure: models hosted locally, agents executing within your perimeter, connected through a hardware-encrypted channel.
No shared tenancy. Processing runs on the device, and in isolated operation no external call is made at all.
- Deployment
- On-premise, inside your perimeter
- Data Residency
- All matter data at rest on your hardware
- Model Execution
- Local inference by default; external models optional and policy-controlled
- Local Agents
- Provisioned from firm-approved templates; scope set by policy, changes logged
- Offline Mode
- Matter work continues offline; externally-sourced modules pause and resume
- Management
- Local console, or remote with scoped, logged and revocable access
- Availability
- Engineering prototype — not yet available for customer deployment
Ready to see Bernio in the context of your firm?
A limited onboarding programme. We configure Bernio for your matter types, practice areas and security requirements.