LEGALIR / Infrastructure

A thin film. A decisive boundary.

Build the thin film between private knowledge and agentic intelligence.

Sign in to learn more ↗
The permissioned thin filmInstitutional records remain in a lower plane. A thin policy boundary lets only permitted retrieval units reach approved agents above. Restricted information stays below.APPROVED AGENTSSCOPED RETRIEVAL ↑POLICY BOUNDARYPRIVATE RECORDSFIG. 03 / PREPARE ONCE. RETRIEVE UNDER AUTHORITY.
PROPOSED INFRASTRUCTURE / IN DEVELOPMENT
01

The interface between systems

LegalIR is designed to sit between approved AI applications and the information an institution authorizes them to use. The gateway receives an authenticated request, resolves its purpose and matter scope, checks source rights, retrieves the minimum relevant evidence and returns source-linked context. The model does not decide its own permissions.

02

Three planes, separate responsibilities

The shared legal graph contains authorized public or licensed authorities. The private data plane contains institution-specific knowledge with tenant, client and matter boundaries. The control plane handles identities, policies, approval requirements and attributable events. A connection between these planes is a governed query, not a license to pool their contents.

03

Deployment follows the data

In-place indexing, customer-managed infrastructure and private inference endpoints are design options. An architecture review must establish where source text, embeddings, prompts, responses, logs and backups are processed. Local storage alone does not establish sovereignty when a downstream model receives confidential context.

04

A request that can be explained

Each retrieval should retain the principal, purpose, effective policy, source versions and decision outcome. A caller receives only permitted units; downstream interpretation must distinguish source evidence from model reasoning. Revocation invalidates future retrieval and triggers removal from derived indexes and caches according to the agreed retention policy.

01 / AI APPLICATION LAYER

The intelligence you choose.

Claude-based agents · GPT-based workflows · Private models
Enterprise legal AI, subject to supported interfaces

↓ Scoped request↑ Evidence-linked response
02 / LEGALIR AGENTIC DATA LAYER

A thin film. A decisive boundary.

Agent GatewayIdentity & PermissionsGraph RetrievalPrivate Retrieval UnitsProvenance VerificationModel RoutingPolicy EnforcementAudit & Usage Metering
↓ Authorized retrieval only↑ Minimum necessary context
03 / CUSTOMER-CONTROLLED ENVIRONMENT

Knowledge stays within its boundary.

Document systems · Matter databases · Internal research
Contracts & opinions · Precedents · Approved data lakes

FIG. 02 — PROPOSED ARCHITECTURE / Boundaries represent access policy, not a claim that model processing never transfers data. Deployment and approved inference endpoints determine the actual data flow.

Your knowledge remains yours.

Enter the reading room ↗