Anonymization engine

Our service is called the VEIL.AI Anonymization Engine.

It is a novel, powerful approach to de-identify personal or otherwise sensitive data, facilitate sharing and analyzing data in low or zero-trust environments and ensure that neither anonymity nor data quality is compromised.

VEIL.AI Anonymization Engine is a versatile tool that can be used for three subclasses of data de-identification. 

  1. It can be used to produce pseudonymized data.
  2. The Anonymization Engine is best used to produce strongly anonymized data. A particular strength of the engine is support for dynamic data sets, for instance, telemetric and sensor data, all continuously updating measurement data, mobility, classroom or finance data.
  3. A particularly interesting use case is synthetic data. This is needed for instance in rapid prototyping, testing of machine learning models, information systems testing and auditing etc.

Ask Tuomo, how our engine works

What our Technology does

DE-IDENTIFIES DATA

VEIL.AI disguises personal identifiers and sensitive attributes of information so that those cannot be re-identified.  These techniques can be applied to small or big datasets, data streams, and volatile data in real-time as it is produced.

BRINGS TOGETHER BIG DATA

VEIL.AI helps you create compatible data sets by harmonizing and anonymizing the data first.  Synthetic data can also be derived to assist in the exploration of viability/usefulness of data sets to new discoveries.

PRODUCES SYNTHETIC DATA

VEIL.AI can produce synthetic data that retains the statistical characteristics of the original data but is completely artificial and carries no re-identification risks.

Our four categories of Sensitive data

SENSITIVE DATA

PSEUDYNOMIZED DATA

ANONYMIZED DATA

SYNTHETHIC DATA

HIGH RISK

MEDIUM RISK

LOW RISK

NO RISK

John – 45 y/o

Original data with personal identifiable information

444555 – Male

Data with all personal identifiers encoded

444555

Data with identifiable information transformed

Max – Male

Statistical data based on real data and data models

Our four categories of Sensitive data

SENSITIVE DATA

PSEUDYNOMIZED DATA

HIGH RISK

MEDIUM RISK

John – 45 y/o

Original data with personal identifiable information

444555 – Male

Data with all personal identifiers encoded 

ANONYMIZED DATA

SYNTHETHIC DATA

LOW RISK

NO RISK

444555

Data with identifiable information transformed

Max – Male

Statistical data based on real data and data models

Read more about our Anonymization engine :