PII Anonymizer
Anonymize personally identifiable information (PII) in your text using our Spacy NER service. GDPR-compliant and privacy-focused.
Original Text
Output format
Clear markers for API integrations; plausible pseudonyms for downstream AI processing (which handles bracket markers poorly).
Anonymized Text
💡 Tip: You can edit this text (e.g. paste LLM output) and click 'Replace Placeholders' to replace [PERSON:a1b2c3d4] etc. with original values.
Statistics
Entities found
0
Text reduction
0%
Detected Entities
The two output modes in detail
We offer two different anonymization formats. Both provide the same protection — the difference is in the output format and therefore in the use case.
Clear markers
Each person, company, email or address reference is replaced by a clearly identifiable placeholder. Format: TYPE plus 8 hex characters in square brackets.
[PERSON:a1b2c3d4] works at [ORG:b2c3d4e5] and can be reached at [EMAIL:c3d4e5f6].
Typical use case: API integrations, automated processing, audit logs. A marker is immediately recognizable as anonymized and can be mapped programmatically.
Plausible pseudonyms
People, companies and addresses are replaced by plausible-looking fake names. The pseudonyms are deterministic — the same name is always mapped to the same pseudonym within a session.
Markus Weber works at Beispiel GmbH and can be reached at markus.weber@example.de.
Typical use case: Further processing by language models (ChatGPT, Claude, Gemini). Pseudonyms read naturally, the model treats them like real names and delivers better results.
Both modes via API: Set the query parameter in your API call
?marker_format=tagged
or
?marker_format=faker.
Your data is processed in real-time and not stored. Session data is cached for 1 hour for de-anonymization purposes only.