US Homeland Security asking after synthetic data systems

Synthetic data – probably including biometric identifiers — has become a must-have for U.S. security officials.

The Department of Homeland Security is looking for private sector products for generating synthetic data to model and simulate databases of real information. The department’s Science and Technology Directorate pushed the solicitation.

Part of the directorate’s Silicon Valley Innovation Program, officials want to create synthetic data at scale, which will enable officials to get the realistic model training they need while minimizing privacy lapses and rights violations.

Participating companies are eligible for as much as $1.7 million in non-diluting investments, according to the directorate. The deadline is April 10.

There are eight capabilities participants must tick. Among them are support for unstructured data types and the artificial generation of data using techniques that reveal insights into the shape and pattern of real data.

The software also must remove or mitigate bias in the artificial data and verify data fidelity. It also has to make it impossible to reverse engineer real data from synthetic.

Biometrics is very current topic for the directorate. It held a biometric technology rally in 2022, with results made public last year, just as it launched a Remote Identity Validation Tech Demonstration.

Article Topics

biometrics  |  biometrics research  |  data privacy  |  DHS  |  RFI  |  synthetic data