Security & Legal Analysis

The legal case for self-hosted legal AI, and the security architecture that makes it possible.

Legal Analysis
Security Architecture
Cloud vs. On-Premises

Security Architecture

Every design decision prioritizes one principle: no data leaves the building.

CounselVault Data Path
Attorney’s Device

Browser on firm network

HTTPS (TLS 1.3)

Local network only

CounselVault

Mac Mini in your office

LLM Inference

On-device GPU

Response

Returns to attorney

CounselVault

Generated locally

No cloud
No API
No training

Only outbound: MDM check-ins (metadata, no user data) and firmware updates (manually approved).

Security Controls

Encryption at Rest

FileVault 2 (AES-256-XTS) enabled at provisioning. Full disk encrypted before shipping. Recovery keys escrowed to MDM.

Encryption in Transit

HTTPS with TLS 1.3 and locally-issued certificate. Traffic never traverses the public internet.

User Authentication

Built-in auth with LDAP/Active Directory. SSO via SAML 2.0 or OAuth 2.0. Multi-factor authentication supported.

Role-Based Access

Admin, Attorney, Paralegal, Read-Only roles. Per-knowledge-base permissions for practice group isolation.

Network Isolation

macOS firewall blocks all inbound except HTTPS on port 3000. Outbound restricted to MDM and updates only.

Air-Gap Capable

The LLM runs entirely on-device via Metal GPU. No API calls during inference. Can operate fully air-gapped.

Audit Logging

Every prompt, response, upload, and modification logged with timestamps and user IDs. Exportable JSON for compliance.

Remote Wipe & Lock

MDM can remotely erase or lock the device. FileVault keys destroyed, rendering all data unrecoverable.

Document Integration

Connects to existing DMS via SMB 3.x or NFS v4. Read-only mounts by default. Supports PDF (with OCR), DOCX, DOC, TXT, RTF, HTML, CSV, Markdown. Documents are never modified — only vector embeddings are created locally.

Cloud AI vs. On-Premises AI

This is not primarily a technology decision. It is a privilege and liability decision.

FactorCloud AICounselVault
Where prompts goTransmitted to provider servers. Subject to retention policies.Processed on Mac Mini in your office. Never leaves local network.
Who can accessProvider employees, automated systems, potentially government.Only authenticated firm users. Full audit trail. No external access path.
Terms of ServiceConsumer: broad data rights. Enterprise: narrower, but still third-party transmission.No ToS. Open-source Apache 2.0. Firm owns the hardware. No AI provider relationship.
Attorney-client privilegeHeppner: consumer not privileged. Enterprise BAA untested under Kovel.No third-party disclosure. Deployed at counsel’s direction. Strongest privilege position.
Work productWarner/Morgan suggest survival, but Heppner’s waiver creates uncertainty.No disclosure to anyone. Work product argument is unassailable.
Model trainingConsumer: your data may train the model. Enterprise: typically opt-out.Never used for training. Model weights are static. No phone-home mechanism.
Cost structurePer-user/month SaaS. AI features are premium add-ons. Multi-year contracts.One-time hardware ($2,999–$7,999) + modest monthly management. Unlimited use.
The Bottom Line

Every major law firm analysis published since Heppner recommends the same thing: enterprise or self-hosted AI with robust confidentiality protections, attorney direction, and no consumer platforms for privileged work. CounselVault is the only productized appliance that satisfies all of these recommendations out of the box.