For Law Firms & Their Clients
Your clients are asking ChatGPT about their case.
Bring that conversation back under counsel.
LitigationVault gives each client a private AI workspace supervised by your firm. Every answer is flagged for attorney review the moment it appears, so the client knows not to rely on it until you’ve approved, edited, or replaced it — and the full conversation stays tied to the matter.
Clients get the AI help they already want. You get oversight, privilege support, and a billable record of the work.
What Your Firm Gets
Built for how your practice actually works.
Your clients are going to use AI. The question is whether those conversations happen inside your firm’s workflow, or somewhere you can’t see, supervise, bill, or defend. LitigationVault gives the client a simple AI workspace and gives the attorney the controls that matter: review on every answer, matter-level records, engagement documents, urgency flags, and billable review time.
Privilege support on every AI conversation
Clients can research their case inside your firm's workspace instead of pasting facts into a public AI tool. The attorney supervises the exchange, reviews every answer the client receives, and keeps the record tied to the representation — the record lawyers need to support privilege if the conversation is ever challenged.
AI review becomes billable work
Reviewing what a client is about to rely on is legal work. LitigationVault tracks the time, ties it to the matter, and gives the firm a new way to capture value from client AI use instead of losing that work to public tools.
One matter record, not screenshots
Every client question, AI response, attorney edit, internal note, approval, replacement, document upload, and timestamp stays in one matter-level record. No forwarded chats. No screenshots. No guessing what the client asked before they called.
Attorney review on every answer
The moment the AI responds, the answer is flagged for attorney review and the client sees it marked as pending. You can approve it, edit it, or replace it — and your changes become what the client sees, labeled attorney-edited. Urgent questions surface first, so a deadline or settlement question never waits behind routine research.
Engagement documents built in
The client signs the engagement and AI-use documents before the first question. That means the relationship, scope, attorney supervision, and AI-use terms are documented at the start instead of reconstructed after a dispute.
Privilege report on demand
If the AI conversation is ever challenged, generate a report showing the client, matter, attorney, prompt, AI response, attorney action, timestamp, and review history — so the firm can explain how the communication was supervised.
How Privilege Is Supported
The elements courts look for — built into the workflow.
“Had counsel directed Heppner to use Claude, Claude might arguably be said to have functioned in a manner akin to a highly trained professional who may act as a lawyer’s agent within the protection of the attorney-client privilege.”
United States v. Heppner, No. 1:25-cr-00503 (S.D.N.Y. Feb. 17, 2026)Start with the relationship
The client signs the engagement and AI-use documents with an assigned, licensed U.S. attorney, and the workspace is tied to that attorney and the matter. That's the attorney-client relationship courts look for first.
Keep the AI inside the vault
The client asks questions and uploads documents inside LitigationVault, not in a consumer AI account. The interaction stays in the firm's workspace and routes through approved enterprise model access (Claude, ChatGPT, or Gemini), so your data isn't used to train the underlying models.
The attorney reviews every answer
Each AI response reaches the client flagged for attorney review, so it's clearly not something to rely on yet. The attorney approves it, edits it, or replaces it — and an edited answer replaces what the client sees, marked attorney-reviewed. This supervision by counsel is the kind of involvement the court in United States v. Heppner pointed to as the path to privilege.
Seal the record
The prompt, response, edits, attorney action, timestamp, and matter metadata are written into a tamper-evident record that can be produced later if the conversation is ever challenged.
Pricing
Simple monthly plans.
Flat monthly pricing by client capacity — custom plans once your firm outgrows the base tier. No per-seat fees and no per-chat surprises.
Standard
$250
per month · up to 20 clients
- ✓Up to 20 active client matters
- ✓Invite clients via email or SMS
- ✓Full attorney review queue
- ✓Clio time-entry sync
- ✓All AI models
- ✓Privilege audit reports
- ✓Engagement letter templates
Custom
Let's talk
for firms beyond 20 clients
- ✓Everything in Standard
- ✓Unlimited active client matters
- ✓Priority support
- ✓Custom firm branding
- ✓Advanced analytics dashboard
- ✓Dedicated account manager
Individual
$50
coming soon
- ✓Assigned U.S.-licensed attorney
- ✓Retainer for vaulting services
- ✓Choose any AI model
- ✓Unlimited vaulted conversations
- ✓Attorney review notifications
Questions firms ask
How does this support privilege?+
LitigationVault documents the pieces lawyers care about: the attorney-client relationship, attorney supervision, confidentiality controls, the purpose of the communication, and the attorney's review of every answer. It doesn't "guarantee" privilege — no tool can — but it gives the firm a far stronger record than an unsupervised consumer AI chat. The court in United States v. Heppner (S.D.N.Y. Feb. 2026) pointed to exactly this kind of attorney involvement.
Why not just tell clients not to use ChatGPT?+
Because many already are. LitigationVault gives them a safer place to ask those questions and gives the firm visibility before the client relies on an answer.
Is the AI giving my client legal advice?+
No. Every AI answer is clearly disclaimed as computer-generated and not legal advice. The supervising attorney can review and edit any response, and every action — the prompt, the AI answer, the attorney's edits, and the approval — is logged to the matter. The legal advice comes from your attorney; the AI is a research tool working under that supervision.
Does this create more work for attorneys?+
It creates review work, but that work is tracked and tied to the matter. The goal isn't to have attorneys review every casual AI interaction forever — it's to capture the client AI use that would otherwise happen outside the firm, where you can't see or bill it.
Does this fit how my firm already bills?+
Yes. Attorney review time is tracked as you work and drafted into time entries that sync to Clio in one click — no reconstructing billable hours from memory.
Which AI models can clients use?+
Claude, ChatGPT, Gemini, and privately hosted models through AWS Bedrock — all through enterprise model access that doesn't train on your data, never consumer apps.
See how a client AI question becomes supervised legal work.
Book a walkthrough and we’ll show a live client question moving through the vault: prompt, AI response, attorney review, client release, and final record. We onboard your firm and its clients by invitation.