Misapplication of Facial Recognition Technology Leads to Wrongful Arrest in the UK

In January 2026, Alvi Choudhury, a 26-year-old software engineer, was wrongfully arrested by Thames Valley police for a burglary that occurred 100 miles away in Milton Keynes. The arrest was based on a facial recognition match that incorrectly identified him as a suspect in a £3,000 theft. Choudhury was at his home in Southampton when police arrived, handcuffed him, and held him in custody for nearly 10 hours before his release at 2 AM. The facial recognition software, procured from Cognitec, a German company, is used by UK police forces to conduct approximately 25,000 searches monthly against a database of around 19 million mugshots.

The Guardian

Choudhury noted significant differences between himself and the suspect depicted in the CCTV footage, stating, “The kid looked about 10 years younger than me,” and highlighted various physical discrepancies. Despite presenting evidence of his whereabouts during the crime, he was taken into custody. Choudhury expressed concern over the implications of being wrongfully arrested again, as he had previously been wrongfully detained in 2021.

The Guardian

Thames Valley police acknowledged that the arrest might have resulted from biases inherent in facial recognition technology, which has been shown to produce higher false positive rates for black (5.5%) and Asian (4.0%) faces compared to white faces (0.04%). A spokesperson for the police stated that the arrest was based on a human visual assessment in addition to the facial recognition match. However, Choudhury's experience raised questions about the reliability of such technology, as he reported that officers at the station laughed when he asked if the suspect resembled him.

The Guardian

Choudhury is now pursuing damages against both Thames Valley police and Hampshire Constabulary, advocating for greater transparency regarding wrongful arrests linked to facial recognition technology. His lawyer, Iain Gould, emphasized the need for police to ensure that artificial intelligence complements human intelligence rather than replacing it.

The Guardian

The incident has sparked renewed scrutiny over the use of facial recognition technology in policing. In December 2024, William Webster, the UK’s biometrics and surveillance camera commissioner, raised concerns about the retention and use of images of individuals who were arrested but not charged. The Home Office is currently reviewing guidance and training to minimize errors in facial recognition applications and is developing a new national facial matching system with improved algorithms.

The Guardian

• “I was very angry, because the kid looked about 10 years younger than me,” — Alvi Choudhury, Wrongfully Arrested Individual

The Guardian

• “must ensure that artificial intelligence is not substituted for human intelligence and due diligence, but instead is used in careful partnership with it” — Iain Gould, Lawyer

The Guardian

• “While we apologise for the distress caused to the complainant in this case, their arrest was based on the investigating officers’ own visual assessment that the individual matched the suspect in CCTV footage following a retrospective facial recognition match, and was not influenced by racial profiling.” — Thames Valley Police Spokesperson

The Guardian

This case underscores the critical need for accountability and accuracy in the deployment of facial recognition technology within law enforcement.

The Guardian