A screen shows a demonstration of SenseTime Group’s SenseVideo pedestrian and vehicle recognition system at the company’s Beijing showroom.

Gilles Sabrie / Bloomberg via Getty Images

This story is part of Generation China, the CNET series exploring the nation’s technological ambition.

Proponents of facial recognition in the United States often argue that surveillance technology is reserved for the greatest risks – to help cope with violent crimes, terrorist threats and human trafficking. And although it is still often used for minor offenses like shoplifting, steal $ 12 worth of goods Where sell drugs for $ 50, its use in the United States still seems tame compared to facial recognition widely deployed in China.

The Chinese facial recognition system registers almost all citizens of the country, with an extensive network of cameras across the country. A database leak in 2019 gave an overview of the ubiquity of China’s surveillance tools – with over 6.8 million recordings in a single day, from cameras positioned around hotels, parks, tourist sites and mosques, recording details about people from 9 days.

The Chinese government is accused of using facial recognition to commit atrocities against Uyghur Muslims, relying on technology to carry out “the largest mass incarceration of a minority population in the world today. ”

“China uses facial recognition to profile Uyghur individuals, classify them on the basis of their ethnicity and distinguish them for tracking, abuse and detention,” a bipartisan group of 17 senators said in a letter to Secretary of State Mike Pompeo on March 11. “And these technologies are deployed in the service of a dystopian vision of technological governance, which exploits the economic advantages of the Internet in the absence of political freedom and sees technology companies as instruments of state power. ”

The aggressive development and use of facial recognition in China provides a window into how technology that can be both benign and beneficial – think facial ID on your iPhone – can also be twisted to enable crackdown. actions that the average person might not even consider criminality. Chinese authorities have used surveillance tools to publicly shame people who wear sleepwear in public, calling it “uncivilized behavior”.

The punishment for these minor infractions is deliberate, surveillance experts said. The threat of public humiliation through facial recognition is helping Chinese authorities direct more than a billion people to what they consider acceptable behavior, from what you wear to how you cross the street.

“The idea is that the authorities try to put in place comprehensive surveillance and behavioral engineering on a large scale,” said Maya Wang, senior researcher on China at Human Rights Watch. “The authorities want to create some kind of society that is very easy for them to manage.”


Now playing:
Check this out:

China turns to technology to monitor, shame and assess citizens


1:58

Behavioral engineering

The idea that China’s facial recognition would automatically be put your name and photo on a billboard for jaywalking and send you a private fine instills fear in people of behaving in a certain way, Wang said.

Technology has often had this ability to affect behavior, even outside of China.

Behavioral engineering is the concept of combine technology and psychology to get people to act a certain way, and it’s something we see every day. It is evident in “dark patterns, “interface designs like hidden unsubscribe buttons that trick people into giving away their personal details, for example.

cnet-china-package-logo-badge-square

Generation China is a CNET series that examines the technological areas in which the country seeks to take a leadership position.

Brett Pearce / CNET

But there is a key difference in how behavioral engineering is done in the United States compared to China and in its facial recognition.

“It’s expressed commercially in the United States, while in China it’s a state effort,” Wang said. “The dynamics are different and the amount of power is different, but there are some striking similarities.”

In the United States, behavioral engineering can be done by amassing data about people and offering or excluding content to them based on predicted personality traits.

Facebook Scandal 2018 with Cambridge Analytica stems from the fact that the now defunct UK data analytics company used data from millions of people to target advertising it would encourage people to vote in a certain way.

Russia’s efforts to interfere with the 2016 U.S. election involved the use of Facebook to create division groups and organize events where people would show up to protest. These positions would be intended for police groups or specifically for Spanish speakers, for example.

While this type of behavioral engineering is primarily aimed at selling products and profits, China’s push is more aimed at instilling fear and control over its people – and facial recognition plays a key role in that.

“It conditions the idea of ​​’this is the way society works’, that you are always being watched,” said Jake Laperruque, senior counsel at the Constitution Project. “It sounds like trying to reinforce to people that” we watch all the time, if you ever do anything to piss us off, we’ll see and we’ll find a way to embarrass you. “”

China’s record

In China, no one is immune to facial recognition or the public shame that comes with it.

Cameras installed at crosswalks to identify and post photos of walkers are commonplace, and a Abacus report in May 2019 showed pictures of children walking on a digital billboard. Local traffic police said “children should be treated the same as adults,” according to the outlet.

A chinese park also capitalized on facial recognition to prevent people from taking too much toilet paper, scanning people’s faces before distributing the rolls.

“It’s kind of a ‘anything goes’ system out there, even if it’s a minor,” Laperruque said of Project Constitution. “Any sort of offense goes for this kind of technology.”

And while in the United States, facial recognition is judged on racial prejudice and human rights concerns, in China, providers of surveillance technology boast of its ability to distinguish people of different ethnicities.

Last November, IPVM discovered that Chinese surveillance firm Hikvision marketed that its cameras could automatically identify Uyghur Muslims with his facial recognition.

There is a setback in facial recognition in the United States because researchers are able to shed light on their concerns about the technology’s racial bias, but there is no similar examination in China, said. Researchers.

China’s facial recognition accuracy rates are not questioned, even when it accidentally detects a person on a bus ad and considers her a thug.

“Americans can talk about racial inequality without fear,” Human Rights Watch’s Wang said. “These systems are neither discussed nor rejected in China. There is no authorized space, both online and offline, to discuss racial prejudice and the persecution of minorities in China.

Recognize the influence

China’s model of facial recognition, from its oppressive surveillance of Uyghur Muslims to its daily grasp over a population of 1.4 billion people, is also of concern to the international community, US lawmakers have said.

Last September, the Carnegie Endowment for International Peace discovered that China was a major supplier of AI surveillance, providing technology to 63 countriesThe next closest supplier of AI surveillance technology was the Japanese company NEC, which sells facial recognition in 14 countries.

In April, Reuters reported that Amazon bought cameras from Dahua, a Chinese surveillance company blacklisted by the United States over claims it is helping China detain and monitor Uyghur Muslims.

The $ 10 million deal was for 1,500 cameras to help monitor the spread of COVID-19, according to the report.

In the senators’ letter of March 11, they raised questions about China’s influence on how facial recognition should be used as it is the technology’s top exporter.

Wang noted that while facial recognition is often a subsidiary of US tech giants, like Amazon’s Rekognition or Microsoft, in China, several companies are already dominating the industry. The country’s acceptance of a surveillance state allows facial recognition providers to advance the technology for even the most trivial of uses.

With this level of influence, there are fears that the Chinese model of using facial recognition – a widespread network designed for public shame and control – could spread to the rest of the world.

“Unfortunately, China has indicated its willingness to use standards bodies in perverse ways to normalize global views on Orwellian surveillance technology,” the group of lawmakers said. “By shaping the debate on the legitimate uses of artificial intelligence and facial recognition, China can expand opportunities for countries, especially those in the developing world, to use Chinese surveillance technology.”