robotnurse.io

./legal-ai — client-confidential AI for law firms

Client-confidential AI that never leaves the office.

AI stopped being optional policy the moment the ABA wrote it down. Formal Opinion 512 makes the ordinary duties of competence, confidentiality, communication, and supervision apply fully to generative AI — and the confidentiality duty is the one that decides which tool you can use. A model on the firm's own hardware satisfies that duty the same way a locked file cabinet does. Built for firms in New York City and nearby New Jersey.

✎

RobotNurse sets up private, offline AI on hardware your firm owns. This page is about legal ethics and AI — not healthcare, nursing, or hospital robotics. Nothing here is legal advice; confirm specifics with ethics counsel.

[ the duty that constrains which AI you can use ]

confidentiality — Rule 1.6 protects information relating to the representation, regardless of source. The question any firm has to answer about an AI tool is: where does that information go, who can see it, and is it retained or used to train a model? A consumer chatbot whose terms allow retention or training cannot receive client-confidential information.
competence — Rule 1.1 a lawyer using AI must understand its benefits and risks to the level needed to use it responsibly, or consult someone who does.
communication — Rule 1.4 clients may need to be informed about how AI is being used in their matter, depending on the circumstances.
supervision — Rules 5.1 & 5.3 partners and managers are responsible for the AI conduct of subordinate lawyers and nonlawyer staff, exactly as if they'd used a fax machine carelessly.
candor — Rules 3.1 & 3.3 AI output must be verified before it is asserted or filed; a hallucinated citation is the lawyer's error.
fees — Rule 1.5 charging for AI-assisted work must still be a reasonable fee for services actually rendered.

[ how a local model maps onto the duties ]

the disclosure question running the model on the firm's own hardware turns “do we trust this vendor's terms” into “do we trust our own server room” — a duty the firm was already satisfying before AI existed. No client information leaves the office, nothing is retained by a third party, nothing trains someone else's model.
a record of what ran a local model with audit logging gives the firm a record of what was run against which data — useful if the supervision or candor duties are ever examined.
the honest trade you exchange a vendor-trust problem for an operations problem: patch management, access control, backups, and physical security become yours. Sizing the box is its own decision, which the RAM calculator handles.

[ what it does not fix ]

Verification is still yours

A local model hallucinates citations exactly like a cloud one. Nothing about on-prem removes the duty to check every fact, citation, and conclusion before it reaches a client or a court.

Competence is still yours

Owning the hardware does not confer understanding. Someone at the firm still has to know what the tool can and cannot do well enough to supervise its use.

Privilege is not a feature of the model

Whether work product or privileged communications remain protected depends on how the tool is used and disclosed, not on where the GPU sits.

Security is now your job

In the cloud, a vendor's SOC 2 report is someone else's problem to maintain. In your server closet, patching, access control, backups, and physical security are yours.

[ the three-question test ]

  1. 01

    Does client information leave the firm?

    If yes, is there a basis under Rule 1.6 — consent, an exception, or a no-training/no-retention agreement? Get it in writing.

  2. 02

    Who verifies the output?

    Name the human and the step. If the answer is “nobody,” the use is not compliant at any price.

  3. 03

    Who is accountable when it's wrong?

    If the answer is “the vendor,” stop. It is the lawyer, always.

A local model makes question 1 easy and does nothing for questions 2 and 3 — a fair summary of where on-prem AI actually helps a law firm.

[ the tax side ]

For firms that also do tax work, there's a separate statutory layer: IRC §7216 requires a client's signed consent before tax-return information is disclosed offshore, and Circular 230 (per OPR Alert 2026-19) requires verifying every AI output. That's covered on the accountants' page →.

CLIENT-CONFIDENTIAL AI FOR YOUR FIRMNYC + NJ

If the answer to question 1 is “yes, and that worries us,” the next step is a conversation, not a purchase. Tell me about your firm and I'll map Opinion 512 and Rule 1.6 onto your actual systems.

book a free scoping call   start with the $899 readiness audit →   see the services →   how local AI setup works →