Amazon investors to vote on selling facial recoginition to police amid widespread backlash

Amazon (AMZN) investors will cast their votes on May 22 to decide whether the e-commerce giant should halt sales of its controversial facial recognition technology to governments and law enforcement agencies around the world.

Amazon’s board of directors, however, is recommending that shareholders vote against the measure, as well as a secondary measure that calls for the company to issue a report on the risks of governments using the technology.

The company already works with law enforcement agencies in the U.S., but the American Civil Liberties Union (A.C.L.U.) and M.I.T. Media Lab have both said the technology is flawed, an accusation Amazon rejects.

Technologies like Amazon’s are facing increasing scrutiny from state and local governments, as well as civil rights groups that fear it will be used to spy on everyday Americans. It has become such a hot issue that companies like Microsoft and Google are rejecting requests for the technologies to be used by some agencies.

The board’s vote

Investors opposed to Amazon’s sale of its facial recognition technology, called Amazon Rekognition, are asking shareholders to vote to: “prohibit sales of facial recognition technology to government agencies unless the Board concludes, after an evaluation using independent evidence, that the technology does not cause or contribute to actual or potential violations of civil and human rights.”

Shareholders are also seeking to have Amazon commission an independent study of Rekognition to determine whether it poses a threat to minorities, immigrants or activists; if it will be sold to authoritarian regimes; and what the operational and financial risks of such issues would cause.

So far, Amazon has piloted its Rekognition technology with the Orlando Police Department in Florida and the Washington County Sheriff’s Office in Oregon. Both instances were met with resistance from civil rights groups, AI specialists, and defense attorneys.

The fear is that Rekognition will disproportionately target minorities, members of civil rights groups, and lead to false arrests.

In July 2018, the A.C.L.U. ran a study that it said matched the headshots of 28 members of Congress to mugshots of known criminals. A secondary test performed by the M.I.T. Media Lab in January 2019 and reported by The New York Times found that Rekognition had a hard time identifying female faces and the faces of dark skinned individuals.

Representatives from Amazon, however, pushed back against those claims, saying both the A.C.L.U. and M.I.T. Media Lab studies didn’t use the Rekognition technology properly. The company also issued a lengthy response statement on how it uses Rekognition.

Lawmakers and other tech companies, though, are calling for greater oversight over the technology.

The response to facial recognition

Ahead of Amazon’s shareholders meeting, the San Francisco Board of Supervisors voted to ban the use of facial recognition technology by law enforcement groups, while Massachusetts currently has a bill seeking to put a moratorium on the tech in committee.

Microsoft (MSFT) President Brad Smith has said that his company rejected the sale of its own facial recognition technology to a police department out of fear that it would disproportionately impact women and minorities. Smith said that the technology had primarily been trained with white males, and, as a result, wouldn’t have been accurate. The company also denied the sale of its tech to a foreign country.

Google (GOOG, GOOGL), meanwhile, has chosen not to sell its technology at all.

In a blog post, Kent Walker, Google’s senior vice president of global policy, wrote, “We continue to work with many organizations to identify and address these challenges, and unlike some other companies, Google Cloud has chosen not to offer general-purpose facial recognition APIs before working through important technology and policy questions.”

Story continues