Simon Mackenzie, a security guard at discount retailer QD Stores outside London, was out of breath. He had just chased three shoplifters who had flown away with several packets of laundry detergent. Before the police arrived, he was sitting at a desk in the back room doing something important: capturing the faces of the perpetrators.
Using an aging desktop computer, he pulled up the footage from the security cameras, pausing to zoom in and save a photo of each thief. He then signed up for a facial recognition program, Facewatch, which his shop uses to identify shoplifters. The next time these people enter a store within a few miles that uses Facewatch, store staff will receive a notification.
“It’s like having someone with you saying, ‘The person you caught last week just came back,'” said Mr. Mackenzie.
use of facial recognition technology was by the police intensively examined in recent years, but its application by private companies has received less attention. As technology improves and its cost decreases, the systems are becoming more and more intrusive in people’s lives. Face recognition is no longer just the purview of government agencies, but is increasingly being used to identify shoplifters, problem customers, and so on legal opponents.
Facewatch, a British company, is used by retailers across the country frustrated by petty crime. For just £250 a month, or about $320, Facewatch offers access to an individual watch list that stores stocks that are close together. When Facewatch detects a tagged face, a notification is sent to a smartphone in the store, where employees decide whether to keep an eye on the person or ask them to leave the store.
Mr Mackenzie adds a new face or two every week, he said, mostly people who steal diapers, groceries, pet supplies and other cheap goods. He said he sympathizes with their economic hardship, but the rate of theft has gotten so out of hand that facial recognition is needed. Usually, Facewatch will alert him at least once a day that someone on the watch list has entered the store.
Face recognition technology is spreading as western countries struggle with advances in artificial intelligence. The European Union is drafting rules that would do this ban many facial recognition applications, while New York City Mayor Eric Adams has encouraged retailers to try the crime-fighting technology. MSG Entertainment, the owner of Madison Square Garden and Radio City Music Hall, used automated facial recognition refuse entry to lawyers whose law firms have sued the company.
Among democratic nations, Britain leads the way in the use of live facial recognition, and courts and regulators have approved its use. The police are here London And cardiff experiment with technology to identify wanted criminals as they walk down the street. In May it was used to scan the crowds at coronation by King Charles III.
However, retail use has been criticized as a disproportionate solution for petty crime. Individuals have little opportunity to know they are on the watch list or how to appeal. In a lawsuit last year, Big Brother Watch, a civil society group, called it “Orwellian in the extreme.”
Fraser Sampson, the UK’s biometrics and surveillance cameras officer, who advises the government on policy issues, said there was “a nervousness and hesitation” about facial recognition technology due to privacy concerns and poorly-performing algorithms in the past.
“But I think that in terms of speed, scope, accuracy and cost, facial recognition technology can literally be game-changers in some areas,” he said. “That means his arrival and deployment are likely inevitable. It’s just a matter of when.”
“You can’t wait for the police to come”
Facewatch was founded in 2010 by Simon Gordon, owner of a popular 19th-century wine bar in central London, known for its cellar-like interior and popularity with pickpockets.
At the time, Mr Gordon hired software developers to develop an online tool to share surveillance camera footage with authorities, hoping it would save police time filing incident reports and lead to more arrests.
Interest was limited, but Mr. Gordon’s fascination with security technology was piqued. He’s been following developments in facial recognition and came up with the idea for a watch list for retailers to share and contribute to. It was like the photos of shoplifters keeping stores next to the checkout but uploading them to a collective database to identify bad guys in real time.
In 2018, Mr. Gordon felt the technology was ready for commercial use.
“You have to help yourself,” he said in an interview. “You can’t count on the police coming.”
Facewatch, which licenses facial recognition software from Real Networks and Amazon, is now available in nearly 400 stores across the UK. Based on millions of images and videos, the systems read a face’s biometric information as the person enters a store and matches it to a database of tagged people.
The Facewatch watch list keeps growing as stores upload photos of shoplifters and problem customers. Once added, a person stays there for a year before being deleted.
“Mistakes are rare, but they do happen”
Each time Facewatch’s system identifies a shoplifter, a notification is sent to the person who passed the test “Super Recognizer” – someone with a special talent for remembering faces. Within seconds, the super recognizer must confirm the match against the Facewatch database before sending an alert.
Although the company has policies in place to prevent misidentification and other errors, mistakes do happen.
In October, a woman buying milk at a supermarket in Bristol, England, was confronted by a staff member and asked to leave the supermarket. She was told Facewatch had flagged her as a prohibited shoplifter.
The woman, who asked for her name to be withheld for privacy reasons and whose story was corroborated by materials provided by her lawyer and Facewatch, said there must be a mistake. When she contacted Facewatch a few days later, the company apologized, saying it was a mix-up.
After the woman threatened legal action, Facewatch combed through his files. It emerged that the woman had been put on watch following an incident ten months earlier involving goods worth £20, about $25. The system “worked perfectly,” Facewatch said.
But while the technology correctly identified the woman, it didn’t leave much room for human judgment. Neither Facewatch nor the store where the incident occurred contacted her to let her know she was on the watch list and to ask what happened.
The woman said she had no recollection of the incident and had never shoplifted. She said she may have left because she didn’t realize her debit card payment at a self-checkout machine had failed.
Madeleine Stone, Big Brother Watch’s legal and compliance officer, said Facewatch was “normalizing airport-style security checks for everyday activities like buying a pint of milk.”
Mr Gordon declined to comment on the Bristol incident.
In general, “mistakes are rare, but they do happen,” he said. He added: “If this happens, we acknowledge our error, apologize, delete all relevant data to prevent a recurrence and offer appropriate compensation.”
Approved by the Data Protection Authority
Civil rights groups have raised concerns about Facewatch, suggesting that its use to prevent petty crime may be illegal under UK data protection law, which requires biometric technologies to have a “substantial public interest”.
The UK Information Commissioner’s Office, the data protection authority, conducted a year-long investigation into Facewatch. The bureau concluded in March that Facewatch’s system permitted by lawbut only after the company made changes to the way it worked.
Stephen Bonner, the bureau’s deputy commissioner for regulatory oversight, said in an interview that an investigation prompted Facewatch to change its policy: it would put up more signage in stores, only giving stores information about serious and violent offenders, and warnings only send out about such repeat offenders. This means people will not be put on watch after a single minor offence, as was the case with the Bristol woman.
“It reduces the amount of personal data stored, reduces the likelihood that people will be wrongly included in this type of list, and increases the likelihood that it will be accurate,” Bonner said. The technology is “not dissimilar to very good security forces,” he said.”
Liam Ardern, the operations manager at Lawrence Hunt, which owns 23 Spar convenience stores that use Facewatch, estimates the technology has saved the company more than £50,000 since 2020.
He called Privacy Risks face recognition exaggerated. The only instance of misidentification he recalled was mistaking a man for his identical twin who had shoplifted. Critics overlook the fact that businesses like his operate on thin margins, he said.
“It’s easy for them to say, ‘No, that’s against human rights,'” Mr Ardern said. If shoplifting doesn’t go down, his stores will have to raise prices or cut staff, he said.
