SARI real-time facial recognition

Released: July 2018

Can you improve this page?
Share your insights with us

SARI ('Sistema Automatico di Riconoscimento Immagini', or Automated System for Image Recognition') is a facial recognition system that enables police forces and sports authorities to identify criminals and suspected criminals. 

Piloted for eight months from 2017 by Italy's national police and introduced in July 2018, the system is based on a database of 16 million mugshots, nine million of which belong to people who have been identified by the police once, while the other seven million are of individuals who have been stopped repeatedly.

SARI has been helpful in the arrests of shoplifters and burglars across Italy. For example, in 2018, it correctly identified two burglars in Brescia. But it also raised concerns about its accuracy and its impact on privacy and legality, largely due to the seemingly high number of mugshots in the police database, which was many more than the police's fingerprint identification system.

SARI Real-time

SARI Real-time is a facial recognition system that uses cameras installed in a particular geographical area and capable of scanning individuals’ faces in real-time. These images are then compared with a government watch-list database of up to 10,000 faces. The database is available to law enforcement upon request.

In addition to concerns about inaccuracies, lawyer Tommaso Scannicchio told ZDNet how real-time systems make people at public events feel under surveillance, causing a 'chilling effect'. 

In April 2021, Italy's privacy regulator the Garante declared SARI Real-time illegal on the basis that it lacked a legal basis for the automated processing of biometric data for facial recognition, and that it would effectively establish an 'indiscriminate mass surveillance system.' 

Operator: Polizia di Stato; Udinese Calcio; S.S.C. Napoli
Developer: SARI Enterprise; Reco 3.26

Country: Italy

Sector: Govt - police; Govt - immigration; Media/entertainment/sports/arts

Purpose: Strengthen law enforcement

Technology: Facial recognition
Issue: Accuracy/reliability; Dual/multi-use; Privacy; Surveillance

Transparency: Governance; Legal; Marketing; Privacy


Research, advocacy

Investigations, assessments, audits

Legal, regulatory

News, commentary, analysis