re:solved, AI automation agency
← Back to blog
Cătălin Andrei Bora, founder of re:solved

By · Founder, re:solved

·9 min read

AI Automation for Law Firms: Scenarios and Boundaries

Hypothetical scenarios for intake, documents and deadlines. We have no law-firm client. Check data processing before construction.

A wall of legal case files and a leather-bound book on a desk beside a small on-premise server rack in a law office at dusk.

Law firms are close to the ideal candidate for AI automation: the work is high-volume, procedural, deadline-driven and expensively staffed. They are also the worst possible candidate for the default approach, because pasting a client matter into a public chatbot is a confidentiality problem before it is a technology problem.

Hosting alone does not establish who sees data. We check access, providers and contracts. Live agents use re:solved’s server in Timișoara and the OpenAI API under a data-processing agreement.


The examples below are hypothetical scenarios, not law-firm client deployments:

  • client intake. Qualifying enquiries, collecting matter details, running conflict checks before a fee earner is involved
  • document triage. Classifying incoming correspondence, disclosure bundles and contracts, then routing them to the right matter file
  • clause and contract review support. Flagging missing clauses, unusual terms and deviations from your own precedent bank
  • deadline and limitation tracking. Chasing the court date, the filing window, the response deadline nobody wants to be the one to miss
  • time-recording prompts. Reconstructing billable narratives from calendar, email and document activity
  • client updates. The routine "where are we" message that clients want weekly and fee earners send monthly

None of this is legal judgement. All of it is the administrative scaffolding around legal judgement, which is exactly the boundary a well-built agent should respect.


Why the cloud-first approach fails a law firm

Everything on that list touches privileged material. When a general-purpose AI tool processes it, the firm has to answer three uncomfortable questions:

  1. Where is the data processed, and by whom? A third-party AI system provider becomes a sub-processor of privileged client information.
  2. Is the material used for training, retention or abuse monitoring? Default consumer terms frequently allow retention windows the firm never disclosed to the client.
  3. Can you evidence the answer? Professional-conduct duties on confidentiality are not satisfied by a vendor marketing page.

Under GDPR the firm remains the controller regardless of which tool the fee earner used, so a well-intentioned paste into a browser tab is the firm's breach, not the vendor's.


The self-hosted alternative

Live agents run in isolated instances on re:solved’s server in Timișoara. Message text is sent to the OpenAI API under a data-processing agreement. Entirely local hosting and processing are design options, not descriptions of live deployments. Local hosting alone does not guarantee compliance.

Live agents run in isolated instances on re:solved’s server in Timișoara. Message text is sent to the OpenAI API under a data-processing agreement. Entirely local hosting and processing are design options, not descriptions of live deployments. Local hosting alone does not guarantee compliance.

ConcernPublic AI toolSelf-hosted legal agent
Where privileged data goesThird-party servers, often outside the EUDepends on hosting and processing
Sub-processor disclosureRequired, and client may objectCheck providers and contracts
Training / retention riskGoverned by vendor terms that can changeRetention agreed contractually
Audit trailWhatever the vendor exposesFull logs the firm controls
Conflict-check dataLeaves the firmDepends on hosting and processing
The deployment AI system, not the AI system itself, is what a confidentiality review turns on.

What a document-triage agent looks like in practice

Hypothetical scenario, not a client: an agent for a practice’s incoming correspondence could:

  1. watches the shared inbox and the scanner output folder
  2. classifies each item. Court correspondence, opponent correspondence, client document, invoice, junk
  3. extracts the matter reference, party names and any date that looks like a deadline
  4. files the document against the right matter in the case-management system
  5. creates the diary entry for any extracted deadline and notifies the responsible fee earner
  6. escalates anything ambiguous to a human with the extracted context attached, rather than guessing

That last step is the one that matters. An agent that refuses to act on an unclear item is behaving correctly; one that quietly invents a matter reference is a liability.


Where the human must stay

  • legal advice and any judgement that carries professional liability
  • final sign-off on filings, undertakings and anything going to a court or an opponent
  • conflict-check outcomes. The agent gathers, a human decides
  • client relationships where the point of the call is that a person made it

The realistic outcome is not fewer lawyers. It is fee earners spending their chargeable hours on the work clients are actually paying for, while the scaffolding runs itself.


How to start without betting the practice on it

  1. Pick one process with a clear trigger and a clear finish. Incoming post triage or new-enquiry intake are the usual winners.
  2. Agree where data is processed and who has access before construction.
  3. Keep a human in the loop on every output for the first few weeks and measure the correction rate.
  4. Extend only after the first process works.

Frequently asked questions

Is it confidential to use AI in a law firm?

Live agents run in isolated instances on re:solved’s server in Timișoara. Message text is sent to the OpenAI API under a data-processing agreement. Entirely local hosting and processing are design options, not descriptions of live deployments. Local hosting alone does not guarantee compliance.

What can legal AI agents automate safely?

Procedural work around the legal work: client intake and qualification, document classification and filing, deadline extraction and chasing, time-recording prompts, and routine client status updates. Legal advice, filings and conflict decisions stay with a qualified human.

Does GDPR allow law firms to use AI on client files?

Yes, but the firm remains the data controller and must have a lawful basis, a processing record, and appropriate safeguards for any processor involved. Self-hosting removes the processor question entirely, which is why it is the simplest compliant route for privileged material.

Can an AI agent do document review for disclosure?

It can triage, classify and prioritise at volume, and flag likely privilege or relevance issues for a human to confirm. It should not make the final relevance or privilege call. That decision carries professional liability and belongs to a fee earner.

How much does AI automation cost for a small law firm?

The 30 minute conversation is free. The specification costs 400 EUR, credited against the build if you sign within 60 days and free with an existing DIGIVEST strategy or feasibility study. A build starts at 1,500 EUR, fixed from the specification. Monthly operation for agents connected to business systems starts at 250 EUR.


Related reading: Is it safe to give your customer data to an AI? (GDPR & self-hosted) · AI automation for healthcare: HIPAA, GDPR and self-hosting · What to look for in an AI automation agency · AI automation services. Managed roles


Want to know which process to automate first?

Book a 30 minute conversation. In 30 minutes we look at one process that costs you time. We tell you honestly whether an agent can take it over and what it would cost. No sales presentation.