Anonymizing means nobody can find out who the person is, not even with reasonable effort. Pseudonymizing means you replace identifying details with codes, such as "Seller A", but a key still exists that links the codes back to the people. Anonymous data is no longer personal data. Pseudonymized data still is.
This page is general information, not legal advice. Follow the guidance of your own chamber or notarial body.
Anonymization in plain words
After anonymization, the link to the person is gone for good. Nobody can bring it back, with the document or with other information they could reasonably get. Under the GDPR, truly anonymous data is not personal data, so its rules do not apply to it.
This is harder than it looks. Removing a name is not enough if the text still says "the only notary in a village of 300 people" or gives an exact date and an unusual amount.
Pseudonymization in plain words
You swap the identifying details for a label and keep a key somewhere safe. The deed now says "Seller A sells to Buyer B", and your own file says who A and B are. This is a good way to work. But the GDPR is clear that pseudonymized data is still personal data, because the key lets you, or someone who gets it, find the person again.
Examples from a deed
| In the deed | Pseudonymized | Anonymized |
|---|
| Jean Dupont | Seller A | the seller |
| 12 Rue des Fleurs, Namur | Property 1 | a house in a town |
| Plot 123/4, price 412,500 EUR | Plot X, price Y | a plot, a price |
| Born 3 May 1961 | Born in year Z | an adult |
In the pseudonymized column, the key still exists. In the anonymized column, you could not rebuild the person, and you may have lost details the work needs.
When is each one enough
- Pseudonymize when you need to put the names back afterwards, for example to get a draft or a translation of a deed and fill the real names in again. Treat the result as personal data and handle it with the same care.
- Anonymize when the text leaves your control, for a training example, a model clause shared with colleagues or a published case study. Then check that nothing is left that points to a person, including the indirect clues.
If you are unsure which one you reached, assume pseudonymized.
Where Acta fits
Acta has a free tool to anonymize a PDF or Word file. The text is removed from the file, not just covered with a black box. You review the result and decide what to keep. Each office's workspace is private, the server is in Germany (Hostinger), and the AI runs through OpenAI's API, which under its API terms does not train on the data. Acta does not train on your documents. For a step-by-step guide, see anonymizing a legal document before AI.
See what Acta can do for your office
Tell me which tasks take your office the most time, and I’ll tailor a free demo to them.
Request a demo