SAN FRANCISCO — Their facial recognition software outperforms Google and China’s best programs — spotting a single face within a crowd of 1 million within a fraction of a second — but the Russian developers behind the technology used by FindFace say what they’ve shown is just the tip of the iceberg.
At a conference called by the National Institute of Standards and Technology, NTechLabs co-founders Artem Kuharenko and Alexander Kabakov showed U.S. government officials what their program is capable of for the first time this week.
"We see applications for this in every field,” said Kuharenko in an interview with BuzzFeed News in San Francisco ahead of the conference. “We are speaking to many companies who are interested in this.”
Kuharenko and Kabakov started NTechLabs in 2015, and promptly created FindFace. They now imagine a world where retailers use FindFace to track the shopping habits of their customers, determining how often they revisit stores and how their shopping habits change over time. Police, they say, can use FindFace to reopen cold cases and search for possible leads, and government security agencies can hunt for suspects on CCTV footage within seconds. They even imagine new possibilities in dating — if you wanted to date someone who looked just like your favorite movie actor or ex-girlfriend, using FindFace could probably get you pretty close.
And yes, they are aware of just how creepy this all sounds.
“You could say our technology really broke private life,” said Kabakov with a nervous laugh. “But really, it is a big question for us, this privacy question.”
In Russia, where FindFace is available for anyone with a profile on Russia’s version of Facebook, VK, privacy activists have already shown how the program is ripe for abuse. In late March, photographer Egor Tsvetkov took photos of random strangers he had snapped on the St. Petersburg subway, and matched them with 70% accuracy to people he found on VK through FindFace. A few weeks later, the Russian version of 4chan used FindFace to find the VK profiles of female Russian porn actors and harass them.
“There are bad guys and they will find bad things to do with it,” said Kabakov, who tried to steer the conversation toward other applications of FindFace. ”In Moscow, officials are using it with thousands of cameras around the city to help stop crime and identify criminals.”
Concerns over how police and state security will use facial recognition software are what keep privacy activists up at night. While facial recognition programs are hardly new, FindFace’s accuracy and speed drew a buzz of excitement when it outperformed Google and Beijing’s FaceAll software last year at the largest annual dataset facial recognition competition, organized by the University of Washington.
The American Civil Liberties Union has already written an ethical framework for facial recognition and in papers has argued that “a technology's intrusiveness must be balanced against the security benefits it would bring. The burden is on the technologists to demonstrate that their solutions will actually be effective in making us safer.”
Kuharenko and Kabakov say it is a question of when, not if, facial recognition becomes a ubiquitous part of public life.
“If we had not done this, improved this program, someone else would have,” said Kabakov. “Everyone is trying to make the best facial recognition program … Everyone sees how useful this will be.”