For law firms

Use AI on your pleadings without the client ever leaving the office.

filterit pseudonymises names, Greek tax ids, social security numbers, IBANs and addresses before the text reaches ChatGPT, Claude or whichever tool you use, and restores them in the answer. Professional secrecy stays in the firm.

Start free See the accuracy

Secrecy has no exception for “the computer”

Whatever you paste into an AI tool is received by a third party, usually outside the EU. The bar code of conduct and the GDPR apply in full.

Pleadings are full of identifiers

Plaintiff, defendant, witnesses, tax ids, id cards, IBANs, addresses, often health data. In three grammatical cases.

“I’ll be careful” does not scale

In a firm with five associates, protection has to be the default of the workflow, not each person’s choice.

What gets detected

What gets detected in your documents

Greek-specific identifiers

  • Tax ID (ΑΦΜ) — Greek tax identification number with modulo-11 check digit validation, not a bare 9-digit match. (check: modulo 11)
  • Social security (ΑΜΚΑ) — Greek social-security number with Luhn and birth-date validation. (check: Luhn)
  • ID card number — Greek police ID card, old and new formats (letters and digits). (check: format)
  • Vehicle plates — Greek registration plates, restricted to the letters Greek plates actually use. (check: format)
  • Court case numbers (ΓΑΚ, ΕΑΚ, ΑΒΜ) — Greek court case and docket numbers, context-gated with NLP. (check: context)
  • Business registry (ΓΕΜΗ) — General Commercial Registry number, 12 digits, context-gated to avoid clashing with other numbers. (check: context)
  • Cadastre code (ΚΑΕΚ) — National Cadastre property code, 12-digit and full 18-digit forms, with prefecture-code validation. (check: prefecture code)
  • Property references (ΑΤΑΚ, ΠΕΑ) — E9 property identifier and energy-performance certificate security numbers. (check: context)
  • Tiresias, auctions, BIC — Greek credit-bureau references, e-auction codes and BIC/SWIFT codes. (check: context)
  • Greek places — Greek gazetteer of cities, municipalities and regions, in every grammatical case.

European and international

  • IBAN — Bank accounts of every EU/EEA country with MOD-97 validation. (check: MOD-97)
  • National IDs (EU, US, UK) — DNI, PESEL, Codice Fiscale, NIR, SSN/ITIN/EIN, National Insurance and more. (check: national checks)
  • Passports — Passport numbers gated by context words in the 24 EU languages (automated tests for 4). (check: context)
  • Driving licences — Driving-licence numbers, context-gated. (check: context)
  • Phone numbers — Landlines and mobiles for the EU, UK, US/Canada and Switzerland, validated per country. (check: libphonenumber)
  • Chassis number (VIN) — 17-character vehicle identification number per ISO 3779, context-gated. (check: ISO 3779)
  • Digital signatures — E-signature identifiers: DocuSign envelope, Adobe Sign agreement, SHA-1/SHA-256 certificate thumbprints. (check: format)

People and contact details

  • Names across grammatical cases — Greek NER that follows inflection, greeklish spellings and links the forms of the same person.
  • Companies and organisations — Company and institution names, Greek and international.
  • Email addresses — Email addresses. (check: format)
  • Street addresses — Street, number and area, in Greek and Latin-script text.
  • Dates — Birth dates and other identifying dates. (check: format)
  • Trademarks — Trademark names from a private list of your own clients.

Health and special categories

  • Medical data — Blood type, record, insurance and prescription numbers, ICD-10 codes, and a condition written behind a caption or as the reason for a sick leave. A condition inside a free sentence usually stays readable.
  • Special categories — Gender, citizenship, religion, marital status and profession written in a captioned form field, and the specialty in staff tables. Inside a free sentence these, like political or trade-union references, usually stay readable.

In scanned PDFs and images, signatures and stamps are covered by visual redaction, with a preview before it is applied.

Every category and the accuracy measurements

How it works, in three steps

Paste or upload

Text, Word, PDF, even a scanned judgment through OCR. You see every finding with its type and correct it by hand.

The AI sees placeholders

“The plaintiff [NAME_1] (tax id [AFM_1]) paid [NAME_2].”. Stable within the document, so the answer comes back right.

The answer returns complete

Placeholders become names again in your own account. The mapping stays encrypted, never with the provider.

Data masking for documents

Preview of findings, export to Word/PDF, restore later. For opinions and long pleadings.

Chat with restore

Type with the real names, the mask goes on automatically, the answer comes back restored. Next to it, the record of what was masked.

The ChatGPT & Claude you already have

A browser extension that masks the moment you press Enter (Chrome, Edge, manual install), or a Windows desktop agent for every application, with central settings, on the Team and Enterprise plans.

Visual redaction on PDFs

Black boxes over data, signatures and stamps, with a preview. The text underneath is removed, not covered.

What you answer the DPO

  • Pseudonymisation under Art. 4(5) and a security measure under Art. 32(1)(a) GDPR.
  • The mapping is encrypted (AES-256-GCM), kept on EU infrastructure (Germany) per account, and purged automatically after 30 days.
  • An audit trail per conversation and an evidence pack for the DPO or the client who asks “what left?”.
  • Dedicated-server and on-premises editions are available by pilot arrangement. A residency policy limiting LLM providers to the EU is in development.
  • A control mapping for ISO/IEC 27701 and 27001 with the evidence for each control, public at filterit.ai/trust/control-mapping, for the DPO file. It supports documentation, it does not certify.

Guides for the GDPR file

AI without leaks: a guide for lawyers

Professional secrecy, GDPR and three ways of working with pseudonymisation. Greek.

Read PDF

GDPR checklist for AI tools in a Greek office

12 points with the articles next to each, a one-page AI-use policy template, what to mask. Greek.

Read PDF

Frequently asked

Do placeholders confuse the AI?

Not in practice. “[NAME_1] paid [NAME_2]” is handled like “A paid B”. Consistency within the document is what matters, and it is guaranteed.

How accurate is it in Greek?

We publish the numbers with every release on the accuracy page, with the gold set and the thresholds. What we measure, we show.

Can I tell the client their data is “anonymised”?

Say “pseudonymised before any AI use, and the provider never receives it”. It is not anonymisation while a mapping exists, and your client may know that.

Try it on a real pleading today

Free account, no card. Data masking works from the first minute.

Start free

Legal references are general information, not legal advice. Accuracy figures are published on the accuracy page and refreshed with every release.