Lacoste Experiments With A.I. to Combat Fashion’s Counterfeit Crisis

By analyzing minute visual details of key product features, such as Lacoste's famous crocodile logo, an A.I. model claims it can detect counterfeit goods with 99.7 percent accuracy.

A guest is seen wearing a white and green Lacoste sweater
Lacoste in October teamed up with Cypheme to deploy an A.I.-powered anti-counterfeit solution across its warehouses. Daniel Zuchnik/Getty Images

In retail, counterfeiting is a $461 billion issue, with pirated products accounting for 2.5 percent of global trade, according to a 2022 report by Europol. Recently, the luxury sports fashion brand Lacoste found itself tangled in a crafty counterfeiting schemewhere scammers would buy genuine items, swap them out with nearly indistinguishable fakes, and then return them for a refund. Facing a surge in returns—especially from online shoppers dealing with size, color and print mismatches—the French brand decided to centralize all returns at its warehouse in Troyes, France where a dedicated team manually investigates each item to ensure it’s authentic.

Sign Up For Our Daily Newsletter

By clicking submit, you agree to our <a href="http://observermedia.com/terms">terms of service</a> and acknowledge we may use your information to send you emails, product samples, and promotions on this website and other properties. You can opt out anytime.

See all of our newsletters

“Counterfeits do more than just tarnish a brand; they pose genuine risks to consumers,” a Lacoste spokesperson told Observer. “These products are more than cheap imitations—they’re often poorly produced using hazardous materials and fabrics that can jeopardize consumer health and safety.”

However, the manual inspection process demanded considerable time and human expertise. And even so, the company has found that some counterfeits are just too good to spot. As a result, in October, Lacoste teamed up with the French tech company Cypheme to deploy an A.I.-powered anti-counterfeit solution, called Vrai AI, across its warehouses. The tool uses advanced visual analysis to spot fake products from just a standard product image—no labels or special markings needed.

By analyzing minute visual details of key product features, such as Lacoste’s famous crocodile logo, Cypheme claims its A.I. model can detect counterfeit goods with 99.7 percent accuracy. “Vrai AI utilizes a proprietary neural network algorithm that can identify a counterfeit it’s never encountered before, a feat that was thought impossible until recently,” the company’s CEO Hugo Garcia-Cotte told Observer.

In addition to advanced image recognition, Cypheme’s proprietary “Noise Print Label” adds an extra layer of protection. This approach uses A.I.-generated, unique digital stickers printed directly on products, making it nearly impossible for counterfeiters to replicate. The technology can not only help verify authenticity but also trace counterfeits back to their sources, offering brands insights into where fakes are originating. Likewise, customers can easily scan a Cypheme-enabled product through a mobile app to confirm its legitimacy at the time of receipt. 

Garcia-Cotte noted the labels also provide real-time tracking across supply chains, which is particularly useful in industries like pharmaceuticals, where knowing a product’s exact journey is critical. “Fake medicine is a critical health crisis in Sub-Saharan Africa, causing an estimated 500,000 deaths annually, with counterfeit antimalarial drugs being one of the leading culprits,” the CEO said. “We’re adapting Vrai AI to authenticate some of the most widely used antimalarial brands in circulation.” By actively collaborating with various NGOs to collect training data and enhance Vrai AI’s algorithm, Cypheme aims to save lives from the devastating effects of counterfeit medicines.

While anti-counterfeit tools like Vrai AI show promise, the battle is far from over. “Our observations show that counterfeiters are constantly finding new ways to create undetectable fakes,” Randall McCarthy, a patent and IP attorney, told Observer. “As A.I. continues to expand, likely counterfeiters will also use A.I. to identify and bypass implemented countermeasures. This will always be an ongoing arms race.”

McCarthy pointed out that, while A.I. is a powerful tool, expecting it to solve the entire global counterfeiting problem is unrealistic. He stressed that industry standardization, collaboration between federal agencies, and the creation of new laws and regulations are essential for unlocking A.I.’s full potential in counterfeit detection. “The problem with counterfeit products goes beyond subpar quality—they can pose life-or-death risks,” he said. “If A.I. systems can be developed to detect these threats, we must do everything we can to make that happen. A.I. systems already outperform human expertise and legal efforts associated with traditional methods.”

Lacoste Experiments With A.I. to Combat Fashion’s Counterfeit Crisis