How anonymize.today Works

Regex-based detection gives 100% reproducible results — the same input always produces the same output. AI and machine learning models can produce different results between runs, suffer from model drift over time, and operate as black boxes that are difficult to audit. For regulatory compliance under GDPR and ISO 27001, organizations need explainable, repeatable processes, which is exactly what regex-based pattern matching delivers.

anonymize.today provides confidence scores from 0.0 to 1.0 for each detection. Users can set minimum confidence thresholds to control sensitivity. Pattern-based entities like credit card numbers and SSNs achieve 95-99% accuracy, while NLP-based entities like names and locations achieve 85-95% accuracy. The platform supports 256 entity types with carefully crafted patterns for each.

Yes, every detection in anonymize.today shows the exact pattern matched, the confidence score, and the entity type identified. The Analyzer highlights detected entities with category-specific colors and positions within the text. This full transparency makes it straightforward to explain detection decisions to auditors, compliance officers, or data protection authorities.

Text submitted to anonymize.today is sent via TLS 1.3 encrypted connections to ISO 27001-certified servers in Germany. The text is processed in memory using Microsoft Presidio, and results are returned immediately. No user content is stored on the servers after processing. Data never leaves the European Union.

anonymize.today supports automatic language detection to identify the primary language of a document. For multi-language texts, users can create custom presets that combine entity types across language boundaries. The platform supports 27 PII detection languages using spaCy, Stanza, and Transformer models, enabling detection of country-specific entities like German tax IDs, French NIR numbers, or Japanese My Number IDs within the same document.