Police Spent £320k on Face ID Tech That Achieved Nothing

British Transport Police (BTP) have spent more than £320k and almost 100 hours of officers’ time on a six-month trial of live facial recognition (LFR) technology in London’s railway stations – yielding just one incorrect face match and no arrests. 

The surveillance tech scans passersby and alerts authorities if it recognises someone on a watchlist. It is supposed to help catch people accused of crimes or who breach court orders.

But a freedom of information request by Liberty Investigates, shared with the Guardian in September, revealed that after scanning more than half a million faces between February and July, the tech only sent one suspect alert, and that turned out to be a false positive. 

Despite this failure, Transport for London (TfL) has given permission for the technology to be expanded across London Underground stations too. 

TfL said the technology could make a vital contribution to tackling violence against the capital’s women and girls, whose travel experiences are too-often “shaped by experiences of sexual harassment and sexual offences”.

The UK’s former biometrics and surveillance camera commissioner, Fraser Sampson, told the Guardian that use of facial ID tech must be proportionate to the need. “Success for the police trying to catch people means people being caught. In that respect the trial doesn’t appear to have been very fruitful,” he said.

Sampson is a non-executive director of facial recognition company Facewatch. which specialises in the technology’s use in shops. Facewatch’s own technology has led to a number of people being falsely accused of stealing by supermarkets. 

Sampson’s appointment at Facewatch after leaving his government job was criticised as an “outrageous conflict of interest”. He said at the time that he “put in place specific measures” to avoid the risk of any conflicts, “however limited that potential might be”. 

Liberty Investigates found that more than half of police forces in England and Wales are using LFR. 

A recent report from parliament’s joint committee on human rights called for more robust laws to address the threat of AI to human rights, and called LFR as a “clear example of risk”.