Amazon shareholders have overwhelmingly rejected a proposal to ban the company from selling its Rekognition system to the police and government agencies.

At the company’s recent annual general meeting, less than three percent of shareholders supported the move, which sought to restrict sales of the real-time facial recognition technology to law enforcement agencies.

In both the US and UK, concerns have been raised by civil liberties campaigners, and by politicians on both sides of the house, at the speed of adoption of real-time facial recognition systems, and their potential use in citizen surveillance and racial profiling.

According to a BBC report, Democrat congressman Jimmy Gomez said, “Shareholders did not not end up passing a ban on Rekognition, and you know what? That just means it’s more important that Congress acts.” Republican congressman Jim Jordan added: “It is virtually unregulated, but I think that frankly that needs to change.”

Last year, the American Civil Liberties Union (ACLU) criticised Amazon over sales to two police forces of the Rekognition system, which can be fitted to officers’ body cameras. The technology enables live citizen surveillance, said the union, and risks discriminating against ethnic minorities because of poor training data, leading to people being misidentified and/or profiled by the police.

ACLU lawyer Matt Cagle said that real-time facial recognition “provides government with unprecedented power to track people going about their daily lives. That’s incompatible with a healthy democracy.”

At this year’s AGM, Amazon said that it knew of such concerns but was not aware of any abuse of the system by police. In its belief, it is up to legislators, not individual companies, to curb the technology’s use.

Nevertheless, a second proposal called on the company to commission an independent study into the privacy risks, and whether Rekognition’s adoption could lead to the disproportionate surveillance of ethnic minorities, immigrants, and political activists. The proposal garnered greater support – 27.5 percent of shareholders – but was still rejected.

The debate about facial recognition is certainly hotting up in government. In May, San Francisco became the first American city to ban police and security services from using the technology. The move also prohibits the use of data gathered by facial recognition systems in the city.

City supervisor Aaron Peskin said, “We have an outsize responsibility to regulate the excesses of technology precisely because they are headquartered here.”

The feelings are shared on this side of the Atlantic. In the UK last year, Parliament’s Science and Technology Committee quoted findings from privacy group Big Brother Watch that the Metropolitan Police had achieved less than two percent accuracy rates with its own automated facial recognition system. In a trial programme, just two people were correctly identified and 102 were incorrectly ‘matched’. The force made no arrests using the technology.

The Committee recommended that such systems should “not generally be deployed, beyond the current pilots” until questions about their effectiveness and potential for bias could be answered.

Politicians and privacy watchdogs are not alone in warning of the potential abuse of these systems. Last year, Microsoft urged the US government to regulate facial recognition.

In a blog post, company president Brad Smith wrote, “Facial recognition technology raises issues that go to the heart of fundamental human rights protections, like privacy and freedom of expression.” He called for “a government initiative to regulate the proper use of facial recognition technology, informed first by a bipartisan and expert commission”.

  • Last year, California introduced data privacy regulations that could form the basis for de facto GDPR-style rules in the US. The California Consumer Privacy Act comes into force in 2020, but was opposed by Google, Facebook, and other advertising-driven technology companies.
  • According to Bloomberg and CNBC reports, the US government is considering adding Chinese technology company Megvii, maker of the Face++ facial recognition system, to a trade blacklist.