Uber Eats driver to amend lawsuit alleging racial bias by app’s biometric verification

A UK judge has ruled that former Uber Eats delivery driver Edrissa Manjang can amend and pursue claims that he was kicked off the ride-sharing app as a result of a racially-biased algorithm in its face biometric system, according to Law360.

The judge had first accepted the amended claims in July 2022 after Uber Eats gave information that contradicted his original claims. She also, however, ruled Manjang can not use emails the company sent him about the system as evidence of harassment. The judge decided the emails were protected by judicial proceedings immunity because the company sent them during litigation and can therefore not be held liable for them. Manjang’s harassment, indirect discrimination, and victimization claims still stand.

A full hearing would be “in the interests of justice… because there has been some inconsistency in the explanation given to the claimant at the time and some lack of clarity surrounding the processes involved in the deactivation,” the judge wrote.

Manjang is Black and of African descent. The app kicked him off in April 2021 after he was unable to complete several biometric checks. Uber Eats claims that the Plaintiff shared his account with others on multiple occasions, which violates the rules of the apps. But the user claims that the app’s facial recognition ID checks resulted in “continuous mismatches,” and he accuses Uber Eats of using a racially-biased algorithm.

Manjang applied to amend his claims three times after Uber Eats shared more on the ID verification process. The company said that the face verification checks still involved humans. This resulted in Manjang bringing a victimization claim, arguing that Uber Eats did not use humans for his checks, despite the fact that they use them for other accounts.

He further claims that deactivating accounts without a final human check was indirect discrimination.

This case is among multiple claims of racial bias in the app’s face recognition algorithm. In 2021, UK drivers of ethnic minority descent claimed that they were either fired or had their accounts blocked after a failed facial verification check on the app.

Moreover, that same year, the App Drivers and Courier Union (ADCU) called on Microsoft, the algorithm’s supplier, to suspend Uber’s use of the face verification. Uber was also ordered to hand over some of its drivers’ data in a Netherlands court case on the matter.

On the other side of the coin, food delivery apps are also under fire for failing to verify the identity of riders as the UK government argues the system enables children to work as delivery riders, according to the Independent.

Immigration Minister Robert Jenrick met with Uber Eats and two other platforms to urge them to eliminate a “substitute” courrier policy that he says is “enabling illegal working,” including by minors. Substitute couriers do not go through the same identity verification process as regular delivery workers on the platforms.

A spokesperson for UK app Deliveroo said that the company has “introduced facial recognition technology which will help to counter any abuse on the platform.”

Article Topics

accuracy  |  biometric bias  |  biometrics  |  face biometrics  |  facial recognition  |  identity verification  |  Uber