Corsight Uses Facial Recognition, AI to Combat ‘Sweethearting’

Corsight has launched a solution that uses facial recognition and artificial intelligence (AI) to combat a form of retail theft in which employees give unauthorized discounts or free items to people they know.

The company offers a module that helps combat this insider fraud — known in the retail industry as “sweethearting” — by using its technology to analyze interactions of employees and customers, identify unusual patterns and flag them for review by the retailer’s security team, BiometricUpdate.com reported Monday (Oct. 28).

The “sweethearting” module is part of Corsight’s larger AI-powered retail security suite that also detects individuals on watchlists, unauthorized access to restricted areas, and loitering that may be associated with theft, according to the report.

In another, separate use of facial recognition technology, it was reported in June that Google was using a security system based on the technology at one of its offices.

In that deployment, the company is using facial recognition technology to spot unauthorized visitors and keep them from entering the office. The system includes interior security cameras that collect facial data and compare it to images from employee badges to spot any unauthorized visitors.

“For many years our security team has been testing and implementing new systems and protections to help keep our people and spaces as safe as possible,” a Google spokesperson told CNBC in June.

In May, Microsoft banned police departments from using its Azure OpenAI Service for facial recognition purposes.

The company added language to its code of conduct saying that this AI service may not “be used for facial recognition purposes by or for police departments in the United States.”

The Azure OpenAI Service enables users to build their own copilots and generative AI applications.

In a deployment of facial recognition technology, eCommerce company Bold Commerce introduced a biometric checkout offering called Bold Checkout in January, saying it lets retailers offer their customers the chance to use their faces and voices to prefill the information needed to complete a purchase.

The company said at the time in a press release that with shoppers expecting a faster checkout, depending on passwords for shopper accounts adds friction and vulnerability to the process.