CORPILUS
Corpilus · Protecting data when you use AI

Use AI with your sensitive data protected.

When someone drafts an e-mail or summarises a contract, client data reaches the AI easily. In supported channels Corpilus recognises sensitive values and, depending on the setting, replaces them, redacts them or warns about the risk.

From practice

AI can help with the task without every personal detail.

Three everyday tasks where the model does not need the sensitive values.

Summarising a client contract

To summarise the terms, the model does not need the client's name or account number.

Drafting a reply to an e-mail

To draft the reply the content of the question is enough; the address and the phone number do not have to go with it.

Working through internal material

Internal documents usually carry personal identifiers and contract numbers that the task itself never needs.

The demo

See what the AI receives.

A simplified demonstration mechanism on fictitious material.

For technicians

Reversible client-side redaction before the prompt is sent. Production engine: 9-language pattern corpus, an optional deep check that also looks for names, and block mode for regulated sectors. This demo uses a simplified pattern set (IBAN, company ID, birth number, e-mail, phone, amounts, names).

The demo runs entirely in your browser and sends nothing anywhere. Even so, please don't paste real sensitive data. This demo runs in your browser and sends nothing anywhere.

What the company gets

Set the rules for using AI right where the work happens.

The team has a clearer way to work with sensitive text. An administrator sets the reaction they want and gets an overview of the events in the protected channels. The scope of the protection follows where you actually use AI.

What the protection recognises
namesnational ID numbers and personal identifierscompany registration and tax IDsIBANs and account numbersemailsphone numbersaddressesamountscontract numberscustom patterns — project codes, internal client names

What gets recognised depends on the type of value and on which check is switched on. Fixed identifiers can be verified by calculation; names and free text are less reliable than that.

How it connects

Where you will use the protection

Supported chat in the browserSupported chat in the browser

Protection works right in the window where the employee types: in ChatGPT, Claude and Gemini.

An extension for Chrome and Edge (Manifest V3). Pattern matching runs locally, with no network request; it activates on sign-in to the workspace and checks nothing while signed out.

A supported AI client and its integration pathA supported AI client and its integration path

Claude Desktop or Cursor can call the protection too, and so can applications that work with AI from their own code.

The MCP server exposes tools for scanning, redacting and anonymising text to Claude Desktop and Cursor; company applications call the same checks through the API.

Your own applicationYour own application

If your application sends requests to a model, Corpilus can sit in between and check what leaves.

A proxy in front of the model; its setup and its behaviour during an outage belong to the integration detail.

Integration options
The limits, in plain words

The protection covers the channels that are connected. A check by patterns and a deep check do not have the same reach. No automatic recognition guarantees that every sensitive piece of information is caught.

Pattern matching runs in the browser extension with no network request. Deep checks and the checks for applications run on the server, so the protection as a whole is not exclusively local.

This protection covers the channel that runs through it. A photographed screen or an unprotected device is something no server-side layer stops — so alongside the technology you also need device protection and a clear agreement inside the company.

How it works

What happens to the data on the wayPseudonymisation, validation and DLP for AI channels

Five questions a lawyer and an IT administrator will both ask, answered without acronyms.Pattern recognition runs locally in the browser extension (Chrome, Edge; Manifest V3), with no network request. A deeper server-side check is optional and off by default.

A label instead of a name, not a blankPseudonymisation, not anonymisation (GDPR Art. 4(5))

A client's name becomes a label and comes back in the answer; the same name gets the same label everywhere, so the model still knows who is meant.

Reversible tokenisation: the same original string gets the same placeholder token within a text, the token → value map is built in the extension, and the answer is de-pseudonymised locally. Because the map can restore the data, the result is pseudonymised, not anonymous — legally it is still personal data.

An invoice number is not taken for a birth numberChecksum after a pattern match

Birth numbers and company IDs are checked by calculation; a string of digits that only looks like one stays unchanged.

Birth number: a valid month (including the offset used for women), a day that exists, and for the ten-digit form divisibility by eleven. IČO: a weighted mod 11 sum. Nine-digit birth numbers issued before 1954 have no check digit, and DIČ and VAT numbers have none at all; those are recognised only by shape and label, so false matches are possible there.

Warn, replace or hold backModes: warn, anonymize, redact and block

The extension can just report a finding, replace the data with labels, black it out, or stop the message. By default it replaces, so work can carry on.

A client-side policy with four modes; anonymize is the default. In block mode a prompt with a finding is not sent and only the cleaned version is offered; if the scan fails, sending stays blocked (fail closed). In the other modes a failed scan does not stop sending.

Only what the answer needs goes to the AIData minimisation, data protection by design, security of processing

To summarise a contract, the model doesn't need the client's name or account number. Recognised identifiers reach it only as labels.

Replacing identifiers before sending is a technical measure for the data-minimisation principle (Art. 5(1)(c)), for data protection by design and by default (Art. 25), and for security of processing, where Art. 32 names pseudonymisation explicitly. Whether the processing as a whole complies is for the controller to assess; the tool does not certify it.

When the AI calls toolsMCP: protection tools and an output filter

Claude Desktop or Cursor call the protection as a tool, and results from the other tools are checked before the assistant sees them.

The MCP server exposes tools for scanning, redacting and anonymising text. Output from the other tools passes through data protection: critical data is anonymised, and if that isn't possible the result is blocked; for high and medium risk the configured mode decides.

For IT professionals

What stays in the browser and what leaves

Pattern matching runs in the browser

The full pattern set ships with the extension, the text never leaves the device for it, and the check works with no network.

Three things call Corpilus

Files you ask it to check, the warning about hidden instructions in pasted text, and the optional deep check that also looks for names. The result comes back immediately; nothing is stored or shared with a third party.

Your own server or NVIDIA DGX Spark

Point the extension at your server's address and even these three checks stay inside your infrastructure.

A pattern corpus in nine languages

Patterns for sk, cs, de, en, es, fr, it, pl and sr cover national identifiers, IBANs, payment cards, phone numbers, emails, postal codes and medical data. Names are found by the optional deep check.

When something fails

A failed scan stops sending only in block mode; in the other modes the text goes through. Chrome updates extensions silently: an open AI tab keeps running the old code until you reload it — the extension detects this and offers to reload the tabs.

Attacks on your website, on logins and on the AI itself belong to the protection of websites and of your own AI.

How Corpilus protection works
Kill switch

One toggle stops outgoing data entirely. Used during an audit or when an incident is suspected.

Questions

Data protection, answered

Does an employee have to change or learn anything?

No. They type into the same chat as before; the browser extension swaps the data before sending and restores it in the reply. No training and no new tool — the extension activates when you sign in to the workspace, and checks nothing while signed out.

Do my texts leave for Corpilus servers?

Pattern matching runs right in the browser: the full pattern set ships with the extension, so the text never leaves your device for it. Only three things call Corpilus — files you ask it to check, the warning about hidden instructions in pasted text, and the optional deep check for names; the result comes back immediately and nothing is stored. With your own server or an NVIDIA DGX Spark, even those checks stay with you.

What does the data protection recognise?

Names, birth numbers and other national identifiers, company registration and tax IDs, IBANs and account numbers, emails, phone numbers, addresses, amounts and contract numbers. Birth numbers and company IDs are also verified by calculation, so an invoice number is not taken for a birth number. Custom patterns — project codes or client labels — you add yourself.

Next step

Let us try the protection on a task your team actually does.

We go through the AI tools you use, the sensitive data involved and a suitable way to connect the protection. Sample material is enough for the first demo.