China Pushing Explicitly-Biased Facial Recognition Standards And Local Tech Companies Are Pitching In To Help

from the [clippy-bmp]-it-looks-like-you're-trying-to-commit-genocide dept

Facial recognition tech is plagued by bias, most of it unintentional. That's why it tends to perform more poorly when attempting to recognize minorities and women. Law enforcement doesn't tend to view these problems as bugs since it, too, operates with many of the same biases. But these are usually the byproduct of faulty inputs, which can be exacerbated by choices made by end users.

In China, the bias is the point. The Chinese government's persecution of its Uighur population has seen local tech companies tasked with providing surveillance tools that single out Uighur Muslims so the government can more efficiently control them.

Huawei is building a system that provides the government with "Uighur alarms" whenever a suspected Uighur passes in front of the government's millions of cameras. According to Huawei, this is still in the testing phase, which means nothing more than there's a plan for it to be put to use. Even if it's no more accurate at identifying Uighurs than it is at identifying criminals, it will likely be good enough to put to real-world use. Collateral damage to innocent residents isn't the sort of thing that slows surveillance rollouts in China.

China wants even more facial recognition power. And it has several partners willing to help it out, even if it means the tech will be used to oppress people, rather than protect people from things like crime and security threats.

China enlisted surveillance firms to help draw up standards for mass facial recognition systems, researchers said on Tuesday, warning that an unusually heavy emphasis on tracking characteristics such as ethnicity created wide scope for abuse.

The technical standards, published by surveillance research group IPVM, specify how data captured by facial recognition cameras across China should be segmented by dozens of characteristics - from eyebrow size to skin color and ethnicity.

"It's the first time we've ever seen public security camera networks that are tracking people by these sensitive categories explicitly at this scale," said the report's author, Charles Rollet.

New standards issued by the Chinese government mandate detection of certain traits, including ethnicity and skin color. These mandates don't just affect the cameras the government owns. They also require this tech be built into cameras operated by private businesses and residential structures.

And it's no longer limited to areas where the Uighur population is largest. It's intended to cover the entire country. The standards issued by the government don't specify which ethnicities the government is most interested in, but other documents seen by IVPM spell it out pretty explicitly.

The police standards above only mention "ethnicity" in the context of China's 56 official ethnic groups. Uyghurs are not explicitly mentioned by these standards, despite the fact that Chinese police often integrate Uyghur-detecting AI software in their security camera networks.

However, a Chinese facial recognition company, Bresee, which is owned by Uniview's parent company TransInfo, uploaded an explainer to its website that "EthicCode" is meant for tracking Uyghurs and Tibetans ('Zang' people in Chinese) specifically.

As is always the case when more details come out about the Chinese government's latest tech company aided affront to humanity, the tech companies named have stepped up to offer very carefully worded non-denials. Here are two US government blacklisted surveillance camera manufacturers (Dahua and Hikvision) not denying they will be contributing to the deployment of tech targeting certain ethnic groups.

Dahua described as "false" media reports that it had helped draft government standards for detecting individual ethnic groups.

"Dahua was not involved in creating the database section of the document that mentions ethnic groups," the company added in an emailed statement.

Asked about the IPVM report, a Hikvision spokesman said the company was "committed to upholding the highest standards and respect for human rights".

"As a manufacturer that does not oversee the operation of our products, we do ensure our cameras are designed to protect communities and property," he added.

Dahua: We didn't write the standards we will be following. Hikvision: Hey, we just make the stuff. Pretty much the upper extent of backbone allowed by an authoritarian government that isn't going to take "no" for an answer. That's why the mandates exist. And if these companies want to continue to exist, they'll do what they're told.

Hide this

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: china, facial recognition

Reader Comments

Subscribe: RSS

View by: Time | Thread

  1. icon
    R.H. (profile), 22 Apr 2021 @ 4:36pm

    Re: Re: Re: mandate all homes to have the gov's cameras in all t

    I remember reading 1984 in high school (about 20 years ago) and no one, from the teacher to us students, seriously thought that anything in there was possible in reality. I guess we were wrong.

Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here

Subscribe to the Techdirt Daily newsletter

Comment Options:

  • Use markdown. Use plain text.
  • Make this the First Word or Last Word. No thanks. (get credits or sign in to see balance)    
  • Remember name/email/url (set a cookie)

Follow Techdirt
Special Affiliate Offer

Essential Reading
Techdirt Deals
Report this ad  |  Hide Techdirt ads
Techdirt Insider Chat
Recent Stories

This site, like most other sites on the web, uses cookies. For more information, see our privacy policy. Got it

Email This

This feature is only available to registered users. Register or sign in to use it.