Name: Sistema de reconocimiento biométrico (in Chile)
Area of Application: Policing
What kind of AI according to the project's official information: Facial recognition
Goals: Facilitate the identification of the perpetrators of a crime when their face has been correctly captured by a camera.
How does it work according to the project's official information: According to Subprefect of Police Muñoz, the system is based on the characteristics of three key elements: cameras, software and database. The audiovisual equipment must comply with basic installation standards (height, contrast, and illumination of the images), the software must comply with the US NIST standard, with accuracy in face recognition higher than 99%. Meanwhile, the database of the Investigative Police of Chile (PDI, Policía de Investigaciones) has a record of more than 250,000 people who have committed crimes in Santiago Metropolitan Region, based on 85 years of investigative experience and the use of biometrics. Since 2016, Microsoft has been digitizing PDI’s databases and all the operations they carry out.
Possible bias: False-positives. Actually, there is strong evidence of false positives with this system (refer to Related links with further information).
Public Institutions Involved: Investigative Police of Chile (PDI) – National Center for Criminal Analysis (CENACRIM, Centro Nacional de Análisis Criminal)
Companies/Private Institutions Involved: Not very clear. Some names are Microsoft and NEC Chile