How AI Is Being Used to Combat Luxury Fakes

by

Originally posted on Racked.com and written by Molly Kendrick

Entrupy and Goat are using AI to find counterfeit handbags and sneakers.

Luxury handbags may exude the confidence of wealth, but take them with a grain of fancy salt — the counterfeit business is booming. Around 2.5 percent of global imports to the US are counterfeit or pirated goods. This amounts to about half a trillion dollars’ worth of merchandise, and these ill-gotten gains fund organized crime.

According to Maysa Razavi, an attorney with the International Trademark Association, shoppers often don’t know that “the same people who are counterfeiting are involved in human trafficking and terrorism.” The next time you don a head-to-toe Gucci ensemble, keep in mind that the clutch you bought for a bargain could be lining the pockets of an underworld kingpin.

In 2017, US Customs seized more than 34,000 shipments of counterfeit goods, an 8 percent increase from 2016. Retailers are scrambling to get ahead of the counterfeiters and have begun to explore how technology can help.

Artificial intelligence has increased the arsenal of tools at their disposal, but AI can be costly to implement. Still, the demand is growing, and a few companies have stopped relying solely on human expertise.

Goat and Entrupy are two very different companies that have developed anti-counterfeit algorithms based on immense databases of information on top luxury brands. They use these databases to look for the tiniest of inconsistencies, ensuring that customers get exactly what they pay for.

Goat is an online platform for luxury sneaker resellers. Luxury sneakers are surging ahead of luxury handbags in popularity, and the boom in counterfeits sneakers has followed. Goat resellers submit photos of their shoes for human experts to analyze before shipping the shoes to Goat for AI authentication. To test the company’s AI, I visited a sneaker store in the Bushwick neighborhood of Brooklyn that online reviewers have accused of selling fake Jordans. I took pictures of a few pairs of their sneakers to submit for testing.

According to Goat CTO Andy Shin, Goat has the “largest and most comprehensive database of sneakers in the world.” He also reports that as time has gone on, the number of fakes that Goat receives has drastically reduced, as its zero-tolerance reputation increases.

For Entrupy, a company that specializes in luxury handbag authentication, I brought two Louis Vuitton handbags to their office in midtown Manhattan — Louis Vuitton is one of the most popular brands in the counterfeit market. I borrowed a genuine Damien Eben Hoxton that a friend had purchased in store, and bought a fake on Canal Street in Chinatown, New York City’s most popular destination for counterfeits.

Before visiting Canal Street, I memorized the specs of a Speedy bag — beige interior, gold hardware, and tan handles with a red trim. I dismissed one I saw outside the subway with the wrong lining and began to fancy myself an expert.

After walking up and down the block, I finally heard the siren call —“Gucci, ma’am?” a man leaning against a storefront asked me. “Louis?” I queried. The man made a phone call, and I followed him to a van where a woman retrieved a Speedy knockoff. “Very nice,” she suggested, opening the bag to show me the Louis Vuitton hardware inside. She asked for $100, but I haggled the price down to $70.

At Entrupy, I met with CFO Will Tan and Devin Battersby, the authentication support and customer experience lead. Before trying out Entrupy’s AI, I asked Battersby to give my fake an initial assessment. She praised the Louis Vuitton branding on the bag’s hardware — counterfeits of a lesser caliber might not have the same attention to detail. I patted myself on the back. Sadly, this was one of the few positive notes Battersby had for my fake.

As someone with years of reseller experience, she isn’t easy to dupe. She noted that the handles had a pinkish hue, instead of the creamy Vachetta leather of authentic Speedy handles. Another problem: There’s no tag with a date code, which we would find hidden in the pocket of a true Louis Vuitton, unless it was a vintage model — “which this,” she said, casting a side-eye at my fake, “does not appear to be.”

But these were merely the observations of an experienced reseller. Entrupy relies on a higher authority, a “convolutional neural network,” which comprises a data set of microscopic images of luxury handbags. To create this database, they first collected bags they believed to be authentic, dating as far back as 80 years.

Once they had enough samples, their algorithm took over, analyzing the tiny details that make up the DNA of the genuine articles. They posit that these details are too difficult for counterfeiters to reproduce. Entrupy’s founders co-authored an article that states, “Even if microscopic details are observed [by counterfeiters], manufacturing objects at a micron or nano-level precision is both hard and expensive.”

Entrupy’s CTO and co-founder, Ashlesh Sharma, completed a PhD at NYU specializing in computer vision, which allows computers to capture an object’s microscopic surface data. Sharma saw the potential to use this technology to map what Entrupy calls the “genome of physical objects.”

Following prompts on the Entrupy app, Battersby took pictures of my fake and entered some basic information about the bag. She clipped her phone to a rectangular box with an attachment that allows her to photograph the bag at around 250 times magnification. “A lot of customers say that this is like a bag ultrasound,” Battersby said, moving the attachment to different parts of the monogram pattern.

Pictures taken, Battersby tapped a button that sent the photos for comparison. My bag had already been flagged because of its lack of a date code, but Battersby tapped the red flag icon to prompt the research team to check again. After just a minute, a small UFO icon appeared on the picture of my bag.

It’s officially “unidentified,” a euphemism that Entrupy uses for fakes. “We don’t like using the word ‘fake,’ because people have emotional attachments to objects,” Battersby explained. She went through the same process with my friend’s Louis Vuitton, and within moments it had received the gold checkmark of authenticity.

In addition to its massive database of authentic bags, Entrupy also studies fakes. Last year, a few members of Entrupy’s research team went to China, where many of the world’s fakes originate. Tan recalls that counterfeiters managed to have counterfeits of the Louis Vuitton x Supreme collection on the Chinese e-commerce site Alibaba before those items became available to the public. How? Was there a leak? Tan is hesitant to offer details. “We’ll just say they’re very savvy.” Can he not tell me? “Probably can’t go too deep into that,” he responded, with the nervous grin of a man who knows too much.

Entrupy’s technology isn’t available to the general public. The company only works with businesses and can’t reveal its biggest clients. Tan speculates that these companies may not want to be associated with even the idea of fakes. The world of luxury goods is shrouded in secrecy, and it seems that even the good guys don’t want to show their hand.

In that same vein, Goat decided not to analyze photos of the suspect Jordans I’d sent, out of concern that any information they give me could help counterfeiters develop better fakes. When I asked a Goat PR rep what their experts look for, he declined for the same reason.

Luxury goods demonstrate economic power, but counterfeiters are chipping away at that prestige every day. Anti-counterfeit technology faces a never-ending battle to stay a step ahead of the counterfeit machine. According to Razavi, the fashion industry needs to be more proactive as well. “People in fashion prefer to talk only among themselves about the true cost of counterfeiting,” she said. To improve the future of retail, “we need to start getting the message out to the public.”