New Lawsearch. Find law. Connect knowledge. — Your legal research, now AI-powered Discover Lawsearch
de fr en

Anonymization

Client data has no place in ChatGPT. 

Our solution finds and replaces personal data in your documents so you can use any AI tool without disclosing personal data.

Runs completely on your computer.

Works locally

Runs entirely on your computer. Your data does not reach any server, not even ours.

Understands your documents

Local AI that understands context. The solution knows that "John Doe" is personal data, but "Tribunal" is not.

Integrated with Claude

Direct integration for Claude Desktop and Claude Code. Send documents directly from the chat to anonymization.

Developed for the legal industry

For everyone who takes data protection seriously.


Law firms

Summarize contracts, draft legal pleadings, analyze case law with AI without disclosing client data.

Courts and authorities

Anonymize files and correspondence before using AI for documentation, summaries, or text suggestions, without endangering the parties' privacy.

Private legal departments

Give your teams a privacy-compliant way to use AI, instead of waging a losing battle against shadow IT.

How it works

The solution detects personal data and automatically replaces it with consistent pseudonyms.


1. Add document

Everything you need to use AI without putting sensitive data at risk.

Developed for professionals who work with sensitive data daily.


Your data never leaves your computer

The solution downloads AI models once and then runs completely offline. Your documents are analyzed by real NLP models on your hardware, not sent to a cloud API. You can verify it: disconnect your internet connection and the solution will continue to work.

Understands context, not just keywords

The solution does not just search for patterns. It understands language. It knows that a date of birth is personal data, but an invoice date is not, and that the name of a client is worth protecting, whereas that of a public court is not.

Restore original data with one click

Use consistent pseudonyms such as [PERSON_1] and [EMAIL_1] throughout the document. Reinsert the AI response. Every name, every address, and every IBAN will be restored exactly.

Works with your existing documents

PDF, DOCX, EML, MSG, CSV, RTF, ODT, XLSX, scanned PDFs and plain text. Open file or paste directly. The solution handles the extraction automatically, including local OCR for scans.

Bank-level encryption

The link between original data and pseudonyms is encrypted with AES-256-GCM. The keys are stored in the secure keyring of your operating system.

Finds what manual checks miss.

Tested on German and English corpora of legal, financial, and HR documents. Pattern recognition finds structured data such as emails and IBANs with over 99% precision. AI recognizes the rest.

We can help you with anonymization.

Get in touch with us. We'll show you the solution and the opportunities it opens.


When would you like to speak with us?
Franz Kummer
Franz Kummer

lic. iur., Owner and CEO Weblaw AG & Weblaw GmbH (Deutschland)