The AI that empowers the law firm

SAUL is an agent designed for law firms, working alongside the lawyer from the case file to the filing. It rebuilds the story, proposes a strategy and shapes the defence, supporting the professional in every choice.

Example conversation with SAUL: the lawyer describes a case, SAUL answers, asks for two details and proposes opening the case file.

Civil

Termination of a supply contract without notice

Termination notified with 15 days' notice instead of the agreed 90; assessment of the loss for the uncovered period.

The case file

Facts, dates and sources: all in order

SAUL reads the documents and rebuilds parties, timeline and legal issues. Every piece of information points back to the document it was drawn from, every provision identified to its source in the official legal databases. It all comes together in an explorable concept map linking documents and the parties involved.

  • Parties

    Who is involved in the case, and in what role.

  • Timeline

    Every fact in date order, once, with all its sources.

  • Sources

    Every provision cited opens on its official source.

  • Case map

    Documents, facts and people linked on a map to explore.

Example of a timeline rebuilt by SAUL: five facts in date order, each linked to the document it comes from; the last one is a calculated deadline.

Timeline

5 facts · 4 documents

  1. Supply contract signed: termination with 90 days' notice

  2. Last regular delivery to the client

  3. The supplier terminates by certified email, with 15 days' notice

  4. Supplies stop

  5. Agreed notice period expires

    Calculated by SAUL: 90 days from 18 Mar 2025

The filings

From strategy to filing

SAUL proposes a reasoned strategy and turns it into an action plan. Each action produces a draft filing. The lawyer reviews it, corrects it, works it through with SAUL's chat and exports the result to Word. The signature stays theirs.

Example: from the case action plan, drafting of a formal notice opens; the lawyer asks SAUL to make a paragraph more forceful and SAUL rewrites it, saving the previous version.

Action plan

From the recommended strategy

  1. Urgent

    Formal notice to the supplier for terminating without the agreed notice

  2. Normal

    Quantify the loss for the uncovered period, 2 April to 16 June 2025

  3. Low

    Assess a claim for damages in court

Pseudonymisation example: in a paragraph of the termination notice SAUL detects two companies and one person, keeps the mapping table encrypted in the firm and sends the model only the pseudonymised version.

Termination PEC.eml

Lawyer's view

What the model reads

Pseudonymised

By this notice, , represented by its legal representative , informs of the termination of the supply contract, effective 2 April 2025.

Confidentiality

Names stay in the firm

Before any reasoning, a detector finds people and companies in every document and SAUL replaces them with consistent pseudonyms across the whole case file. The mapping table stays encrypted in the firm's space: the AI model works on the facts, the lawyer sees the real names.

  • Automatic detection

    People and companies found in every uploaded document.

  • Consistent pseudonyms

    The same name has the same pseudonym across the whole case file.

  • Encrypted space

    Documents and mappings in an AES-256 volume, one per firm.

For every case

Every case has its own agent

For every case file SAUL creates a dedicated agent, specialised in that case, which answers like an experienced trainee. The lawyer sets the practice area and the type of engagement, litigation, opinion or ongoing advice, and the agent reasons accordingly.

Five cards, one per case: for each, the dedicated agent with its name, the practice area, the type of engagement and a line from the agent to the lawyer.

Practice areas

  • Civil
  • Criminal
  • Corporate and commercial
  • Employment
  • Family and succession
  • Tax
  • Cybersecurity and blockchain

The next case can start here