Legal

AI for the day-to-day work of a law firm, without client files leaving your perimeter.

The data room lands on Friday with four thousand documents and the closing is Thursday, and the junior has spent three days reading change of control clauses. Inferana reviews that volume inside your perimeter, with every answer anchored to its document.

The rules that apply to you, point by point

What each regulation in your sector requires, and which part of that the deployment solves.

Legal professional privilege

Royal Decree 135/2021, Spanish Bar Statute
What it requires

An absolute duty of confidentiality covering everything the client entrusts to the firm.

How we solve it

Documentation is processed inside your perimeter, with no prompts stored and access restricted per user and per team.

GDPR

Regulation (EU) 2016/679
What it requires

Minimisation, confidentiality and a legal basis for any transfer of personal data.

How we solve it

Inference in the EU or on your own infrastructure: prompts and responses are never stored or used to train models.

AI Act

Regulation (EU) 2024/1689
What it requires

Systems assisting the administration of justice fall under Annex III as high risk.

How we solve it

Open models with published model cards and licences, access control and a record of every query to document human oversight.

Anti-money laundering

Directive (EU) 2015/849
What it requires

Firm due diligence on certain transactions and retention of the supporting documentation.

How we solve it

Extraction and review of the case file inside your perimeter, with an auditable record of every access.

Your most common use cases

The same ones you already use, specific to your sector.

See all use cases

Every closing ends in the same scene: someone reading documents at midnight.
And the next day, again.

Nobody puts it in the fee proposal, but those hours exist and they come out of the matter's margin. What you hear in the room is always the same:

What happens today

Four thousand documents in the data room and the closing is Thursday.

Corporate partner

My junior has spent three weeks reading change of control clauses.

Senior associate

I billed six hours looking for an opinion already in our archive.

Knowledge management lead

Uploading a client file to that website keeps me awake at night.

Managing partner

Clients now ask in the tender where their documentation is processed.

Firm secretary

I cannot show who opened which document on which matter.

Compliance

My team pastes clauses into a public chat to summarise them.

In-house legal

With Inferana

Review and extraction across the thousands of data room documents in parallel, with every finding anchored to its source.
Clauses are located and compared in one pass, and the junior reviews what comes back flagged.
Search across the firm's own opinions and filings, returning the document and the paragraph that answer.
The file is processed inside your perimeter, on European infrastructure or your own, and prompts are never stored.
You can answer the tender with the exact place of execution and the deployment documentation.
An auditable record of every query, with the user and the team behind it, exportable on request.
An approved tool inside the firm, with frontier models reachable only where you enable them.

Sector FAQs

Wherever you decide: on European infrastructure we manage, or on-premise on the firm's own infrastructure. In both cases prompts and responses are never stored or used to train models, so the file does not sit anywhere outside your control.

With access control per user and per team: each person queries what their profile allows, and it is logged. The firm defines the barriers with its own matter allocation rules. We provide the access mechanism and the audit trail to show they held.

The firm controls the life cycle of the file. Prompts and responses are never stored on our side, and anything you indexed for retrieval is exported or deleted when you decide. With an on-premise deployment, deletion happens on your own infrastructure and under your retention policies.

No. The models are open and arrive already trained. Neither your prompts nor the responses are used to train or improve them. Anything you index for search stays in your deployment and can be deleted whenever you want.

Annex III classifies systems assisting the administration of justice as high risk. Classifying your use and taking on those obligations is the firm's responsibility. We provide open models with a published model card and licence, access control and a record of every query, which is what you need to document traceability and human oversight.

Yes, with the same key. There the query does go out to the model provider, so client documentation is handled with the open models inside your perimeter. Access is configured per team and is logged the same way.

The service and deployment documentation: where inference runs, which dependencies it has and how access is controlled, plus the query log for the matter. With an on-premise deployment, what gets audited is your own infrastructure, under the procedures you already have.

The models are open and the API is OpenAI-compatible: you can take the weights with you and point the code elsewhere by changing a key. Your documents and indexes are exportable. Your exit plan does not depend on us still being around.

Tell us what you want to deploy and we will show you where Inferana fits.

In the demo we go through your case: which models you need, where they run and what it takes to meet the regulation that applies to you.

We reply within one business day.