Anonymization vs. pseudonymization (GDPR): the difference
Anonymization removes the link to a person irreversibly. Pseudonymization replaces identifying data with placeholders and stays reversible with separately held information. What that means for using AI.
Updated
The difference: anonymization changes data so that no link to a person can be established anymore. Pseudonymization replaces identifying details with placeholders but can be reversed using separately held information. Under the GDPR, pseudonymized data is still personal data; fully anonymized data is not.
Anonymization
Anonymization removes the link to a person permanently, so that attribution to an individual is impossible or possible only with disproportionate effort. Under Recital 26 of the GDPR, such anonymous data is no longer personal data, and the GDPR does not apply to it. The price: the process is irreversible, and the original values are gone.
Pseudonymization
Pseudonymization is defined in GDPR Art. 4(5): data is processed so that it can no longer be attributed to a person without additional information, which is kept separately. The link is therefore reversible if you hold the key. Pseudonymized data still counts as personal data and remains subject to the GDPR.
What this means for using AI
For an AI workflow, pseudonymization is usually the right choice, because you want to map the AI’s output back to your real data. That is exactly how Stript works: identifying details are replaced with placeholders, and the mapping is stored locally and encrypted on your device.
The key point: you keep the key. The AI provider receives only the version with placeholders and cannot attribute it to people without your separately held information. You then restore the originals locally afterward.
An important note on residual risk
Stript replaces direct identifiers such as names, addresses, IBANs, or tax numbers. Indirect details in the context, for example a rare combination of profession, location, and date, can in some cases still allow a conclusion about a person. Judging whether a residual risk exists in a specific case stays with the human. That is why, in Stript, you confirm every detection yourself.
Pseudonymize locally for safe AI use. Download Stript for free →