January 6, 2026
mast3r/Depositphotos.com
Should Facial Recognition for Theft Get Another Look?
From Wegman’s in New York City to Sainsbury’s in London — and Foodstuffs North Island in New Zealand — grocers continue to trial facial recognition technologies in the hopes of reducing shoplifting. This comes despite continued outcry around what’s seen as an invasive technology coming from some consumers.
Those employing the technology also cite increased incidents of violence from customers.
National Retail Federation’s 2025 “The Impact of Retail Theft and Violence Report” that came in October found incidents of shoplifting and merchandise theft increased 19% in the U.S. from 2023 to 2024, further compounding the 26% growth observed from 2022 to 2023. In the 2024 survey, 73% of respondents said shoplifters had exhibited heightened levels of aggression and violence, and those incidents worsened in the 2025 survey.
In the most upfront test, Sainsbury’s in September rolled out an eight-week facial recognition technology trial at two stores. In a statement prior to the launch, the U.K.’s second largest supermarket chain cited a survey from a major trade union in the U.K. that found 77% of shop workers have experienced verbal abuse — while more than half (53%) were threatened by a customer.
To protect privacy, only those with a record of criminal behavior are flagged, and all other data generated by the software is instantly deleted.
Simon Roberts, Sainsbury’s CEO, said, “We understand that facial recognition technology can raise valid questions about data and privacy. This trial and subsequent roll out is not about monitoring colleagues or our valued customers. It’s focused solely on identifying serious offenders who have committed acts of violence, aggression, or theft, helping our teams prevent further harm.”
Theft and Privacy Concerns Collide in Retail Settings as Consumers, Retailers Debate Pros and Cons
Privacy group BigBrotherWatch, in a response to the trial, urged government officials to tighten laws against facial recognition. The group said, “Sainsbury’s decision to trial Orwellian facial recognition technology in its shops is deeply disproportionate and chilling. Facial recognition surveillance turns shoppers into suspects, with devastating consequences for people’s lives when it inevitably makes mistakes.”
This past June, the heads of major retailers in New Zealand issued a joint statement saying facial recognition technology is a “powerful and effective tool” alongside other crime prevention resources, such as security guards, fog cannons, body cameras, CCTV, and other tech solutions. The statement followed a promising test of the technology in 2024 by Foodstuffs North Island.
“We are firmly of the opinion that [facial recognition technology], when used fairly and accurately, can be a valuable intervention to help keep customers and employees safe,” they said in a statement.
In New York City, Fairway supermarkets in 2023 began posting signs letting consumers know stores collect or may share biometric information, including eye scans and voiceprints, to combat rising theft levels. Manny Gomez, president of MG Security Services, told CBS News at the time, “This is a step in the right direction, in my opinion. If you have nothing to hide, then you have nothing to hide.”
Last week, Wegman’s began posting similar signs indicating its two stores in New York City were collecting biometric information “to protect the safety and security of our patrons and employees,” according to Gothamist.
Nearly all of the responses in the comments section of the Gothamist piece felt the technology spurred privacy concerns, with many vowing never to shop at Wegman’s again. One said, “This is a complete non-starter for me. Will no longer shop there, no matter what.”
BrainTrust
David Biernbaum
Founder & President, David Biernbaum & Associates LLC
Lucille DeHart
Principal, MKT Marketing Services/Columbus Consulting
Christopher P. Ramey
President, Affluent Insights & The Home Trust International
Recent Discussions
Discussion Questions
Why are retailers still exploring the use of facial recognition tech to deter shoplifters despite significant backlash from privacy groups and consumers in trials?
Can enough guardrails be established to lessen privacy concerns?
Poll
Loading …
Why do the efforts continue? Probably because theft hasn’t stopped. 🙂
Do i expect either of these to end? No.
We’ll see more efforts, and more anti-effort legislation…particularly as mistaken ID cases grow.
I am an opponent of the using facial recognition for marketing or any other purpose other than loss prevention and security. I understand that some consumers will be offended and choose to shop elsewhere. That’s their prerogative. However, retail operators have the right to protect themselves against theft and obligation to maintain a safe environment for customers and staff. Given the sharp increase in retail theft, I believe that it’s completely reasonable and appropriate to carefully apply this technology, but only for loss prevention, safety, and security.
Facial recognition works for loss prevention, but in an opt-in basis. “Known shoppers” can be recognized and have reduced security setting at checkout (or perhaps with locked cabinets), but security remains tight for the rest of us. This helps frequent shoppers expedite their trips, and frees up store associates to focus on other tasks and not spend as much time assisting them.
Facial recognition technology can help retailers significantly reduce theft by quickly identifying known shoplifters, thus enhancing security. Additionally, it can streamline the shopping experience by identifying loyal customers and offering personalized services or promotions. The technology also aids in efficient inventory management and loss prevention, ultimately contributing to increased profitability.
In theory, facial recognition sounds like a great tool against (widespread) theft. The problem is that existing products frequently mis-identify people of color. I’ll be excited about using it when it’s ready.
The optics aren’t right yet. We’ll get there. The narrative will be closer to protecting and serving customers rather than stopping shoplifters.
I think stronger legal implications for shoplifters would be a better deterrent than facial recognition. This type of technology, while valid, is too intense for a retail venue. Leave this for airports, banks and federal buildings. I applaud the innovation around finding solutions, but I do not like the concept of facial recognition. Even with auto/instant deletion, this opens up a new level of consumer vulnerability.
Retailers are still exploring facial recognition because theft has become a safety issue, not just a loss issue. Even with guardrails, there will always be some people who are uncomfortable with it, but right now public acceptance is still largely negative. Until that shifts, retailers will be forced to weigh trust against control.
Retailers are still exploring the use of facial recognition technology not because they relish debate with privacy advocates, but because they’re confronting real economic stress points — particularly in urban markets where shoplifting and organized retail crime have driven store closures, excessive shrink, and heavy security spend on locked-up merchandise that suppresses sales. Traditional loss-prevention tactics only go so far, and many retailers feel the pressure to innovate when their very ability to operate profitably in underserved neighborhoods is at stake. In that context, advanced tech like facial recognition — when thoughtfully implemented — can potentially deter theft, enhance employee safety, and keep stores open and inviting rather than shuttered or fortress-like. The backlash from privacy groups and some consumers is understandable, but it doesn’t negate the underlying business imperative to protect assets and jobs in areas where conventional loss prevention has proved inadequate.
At the same time, the legitimacy of those privacy concerns must be taken seriously. Yes, retailers should continue to test and learn with this technology, but not in a vacuum or without strong guardrails. Clear policies around data retention, opt-in/opt-out mechanisms, transparent signage, third-party oversight, and strict use-case limitations (e.g., only for loss prevention and not for marketing or behavior profiling) are essential to establish trust with customers and communities. Deployments should also be paired with independent audits and strong legal compliance so that the use of the technology doesn’t erode the very goodwill retailers are trying to build. Thoughtful communication — explaining why the tech is being piloted, how data is protected, and what safeguards exist — can go a long way toward balancing safety and privacy.
If facial recognition can meaningfully help stores stay open and profitable in communities that have been underserved or abandoned because of theft pressures, then it’s certainly worth responsible exploration. The alternative — stores that lock up merchandise, reduce service, or close entirely — harms both consumers and employees. A test-and-learn approach that pairs innovation with robust privacy protections and ongoing industry dialogue offers the best path forward: researching solutions that protect people and property, without compromising the fundamental privacy rights that consumers rightly expect.
It’s so easy to say, “if you have nothing to hide, then you have nothing to hide” when technology isn’t aimed at you. But that aside, how effective is it really? A professional makeover for headshots was enough to defeat my iPhone’s facial recognition.
This is a capital allocation problem. Facial recognition solves an operational problem (theft), but the real battle is whether retailers build direct value propositions that consumer agents recognize as genuinely beneficial, or become interchangeable fulfillment points optimized for metrics that no longer drive consumer decisions. Retailers are missing the strategic positioning question: how to remain relevant in agent-mediated commerce where consumer-controlled agents evaluate and transact on their behalf. Retailers ought to be asking, “What makes our stores worth choosing when rational agents, not manipulable humans, control purchase decisions?”
If major chains like Rite Aid and Home Depot couldn’t get the ethics right, why should we trust that the current trials will be any different?
Using the technology for security purposes is fine, although retailers need to be careful as instances of misidentification can cause real problems. However, while identify shoplifters and thieves is one thing, getting the criminal justice system to prosecute and punish them is quite another.
Drinking my morning coffee at my regular coffee spot, writing RetailWire as usual, I look up at the camera on the wall.
Please. Can we get over this “privacy” thing? Now hear this… there is no such thing as privacy…anymore. It is something to be concerned about, to complain about, to talk about, but we have given up our privacy a long time ago. If you think privacy is intrusive now, wait a few more years.
One of the most effective ways to deter a shoplifter is to intervene when they’re considering stealing. Facial recognition gives the store the opportunity to greet a known bad actor by name when they enter. Friendliness aside, it lets the person know that “we see you and we know who you are.” This raises the risk factor significantly and often results in that person turning around and walking out.
However, using this technology to identify customers to offer them “surveillance pricing” or send them personalized messaging. That… is a new level of creepy that’s going to cause significant friction between consumer advocates, privacy advocates, government regulators, and retailers. I suspect I may not live long enough to see dust settle on that debate.
The technology has it uses and benefits. NYPD reported last year that they had been able to find a number of vulnerable elderly adults who were “lost” on the street using this tech. There have also been numerous reports of people being identified as criminals solely on the basis of physical characteristics. We are early on in this discussion.
What stood out this holiday season was not the technology itself, but how it blended into the shopping experience. Customers used AI tools to find products faster and get quick answers when issues came up, especially around shipping and returns. When it works like that, it feels less like new tech and more like a natural part of shopping.
From a retail perspective, this only succeeds when the fundamentals are in place. Accurate inventory, reliable fulfillment and clean handoffs between online and stores still matter most. AI can make shopping smoother, but it cannot fix broken processes.
More Discussions
Will In-Store Media Fly For WHSmith?
Is WHSmith’s gamble a wise one? Will broader in-store retail show growth moving forward?What Lessons Can Be Learned From Eddie Bauer’s Retail Exit?
What can be learned from the Eddie Bauer story?Is High Exec Turnover Hurting Retailers?
Is the high degree of executive turnover in retail hurting business?
Why do the efforts continue? Probably because theft hasn’t stopped. 🙂
Do i expect either of these to end? No.
We’ll see more efforts, and more anti-effort legislation…particularly as mistaken ID cases grow.
I am an opponent of the using facial recognition for marketing or any other purpose other than loss prevention and security. I understand that some consumers will be offended and choose to shop elsewhere. That’s their prerogative. However, retail operators have the right to protect themselves against theft and obligation to maintain a safe environment for customers and staff. Given the sharp increase in retail theft, I believe that it’s completely reasonable and appropriate to carefully apply this technology, but only for loss prevention, safety, and security.
Facial recognition works for loss prevention, but in an opt-in basis. “Known shoppers” can be recognized and have reduced security setting at checkout (or perhaps with locked cabinets), but security remains tight for the rest of us. This helps frequent shoppers expedite their trips, and frees up store associates to focus on other tasks and not spend as much time assisting them.
Facial recognition technology can help retailers significantly reduce theft by quickly identifying known shoplifters, thus enhancing security. Additionally, it can streamline the shopping experience by identifying loyal customers and offering personalized services or promotions. The technology also aids in efficient inventory management and loss prevention, ultimately contributing to increased profitability.
In theory, facial recognition sounds like a great tool against (widespread) theft. The problem is that existing products frequently mis-identify people of color. I’ll be excited about using it when it’s ready.
The optics aren’t right yet. We’ll get there. The narrative will be closer to protecting and serving customers rather than stopping shoplifters.
I think stronger legal implications for shoplifters would be a better deterrent than facial recognition. This type of technology, while valid, is too intense for a retail venue. Leave this for airports, banks and federal buildings. I applaud the innovation around finding solutions, but I do not like the concept of facial recognition. Even with auto/instant deletion, this opens up a new level of consumer vulnerability.
Retailers are still exploring facial recognition because theft has become a safety issue, not just a loss issue. Even with guardrails, there will always be some people who are uncomfortable with it, but right now public acceptance is still largely negative. Until that shifts, retailers will be forced to weigh trust against control.
Retailers are still exploring the use of facial recognition technology not because they relish debate with privacy advocates, but because they’re confronting real economic stress points — particularly in urban markets where shoplifting and organized retail crime have driven store closures, excessive shrink, and heavy security spend on locked-up merchandise that suppresses sales. Traditional loss-prevention tactics only go so far, and many retailers feel the pressure to innovate when their very ability to operate profitably in underserved neighborhoods is at stake. In that context, advanced tech like facial recognition — when thoughtfully implemented — can potentially deter theft, enhance employee safety, and keep stores open and inviting rather than shuttered or fortress-like. The backlash from privacy groups and some consumers is understandable, but it doesn’t negate the underlying business imperative to protect assets and jobs in areas where conventional loss prevention has proved inadequate.
At the same time, the legitimacy of those privacy concerns must be taken seriously. Yes, retailers should continue to test and learn with this technology, but not in a vacuum or without strong guardrails. Clear policies around data retention, opt-in/opt-out mechanisms, transparent signage, third-party oversight, and strict use-case limitations (e.g., only for loss prevention and not for marketing or behavior profiling) are essential to establish trust with customers and communities. Deployments should also be paired with independent audits and strong legal compliance so that the use of the technology doesn’t erode the very goodwill retailers are trying to build. Thoughtful communication — explaining why the tech is being piloted, how data is protected, and what safeguards exist — can go a long way toward balancing safety and privacy.
If facial recognition can meaningfully help stores stay open and profitable in communities that have been underserved or abandoned because of theft pressures, then it’s certainly worth responsible exploration. The alternative — stores that lock up merchandise, reduce service, or close entirely — harms both consumers and employees. A test-and-learn approach that pairs innovation with robust privacy protections and ongoing industry dialogue offers the best path forward: researching solutions that protect people and property, without compromising the fundamental privacy rights that consumers rightly expect.
It’s so easy to say, “if you have nothing to hide, then you have nothing to hide” when technology isn’t aimed at you. But that aside, how effective is it really? A professional makeover for headshots was enough to defeat my iPhone’s facial recognition.
This is a capital allocation problem. Facial recognition solves an operational problem (theft), but the real battle is whether retailers build direct value propositions that consumer agents recognize as genuinely beneficial, or become interchangeable fulfillment points optimized for metrics that no longer drive consumer decisions. Retailers are missing the strategic positioning question: how to remain relevant in agent-mediated commerce where consumer-controlled agents evaluate and transact on their behalf. Retailers ought to be asking, “What makes our stores worth choosing when rational agents, not manipulable humans, control purchase decisions?”
If major chains like Rite Aid and Home Depot couldn’t get the ethics right, why should we trust that the current trials will be any different?
Using the technology for security purposes is fine, although retailers need to be careful as instances of misidentification can cause real problems. However, while identify shoplifters and thieves is one thing, getting the criminal justice system to prosecute and punish them is quite another.
Drinking my morning coffee at my regular coffee spot, writing RetailWire as usual, I look up at the camera on the wall.
Please. Can we get over this “privacy” thing? Now hear this… there is no such thing as privacy…anymore. It is something to be concerned about, to complain about, to talk about, but we have given up our privacy a long time ago. If you think privacy is intrusive now, wait a few more years.
One of the most effective ways to deter a shoplifter is to intervene when they’re considering stealing. Facial recognition gives the store the opportunity to greet a known bad actor by name when they enter. Friendliness aside, it lets the person know that “we see you and we know who you are.” This raises the risk factor significantly and often results in that person turning around and walking out.
However, using this technology to identify customers to offer them “surveillance pricing” or send them personalized messaging. That… is a new level of creepy that’s going to cause significant friction between consumer advocates, privacy advocates, government regulators, and retailers. I suspect I may not live long enough to see dust settle on that debate.
The technology has it uses and benefits. NYPD reported last year that they had been able to find a number of vulnerable elderly adults who were “lost” on the street using this tech. There have also been numerous reports of people being identified as criminals solely on the basis of physical characteristics. We are early on in this discussion.
What stood out this holiday season was not the technology itself, but how it blended into the shopping experience. Customers used AI tools to find products faster and get quick answers when issues came up, especially around shipping and returns. When it works like that, it feels less like new tech and more like a natural part of shopping.
From a retail perspective, this only succeeds when the fundamentals are in place. Accurate inventory, reliable fulfillment and clean handoffs between online and stores still matter most. AI can make shopping smoother, but it cannot fix broken processes.