Uber’s automated dismissals – a sign of times to come?

• Considered as the biggest case on Article 22 of the GDPR to reach the courts, a group of former Uber drivers hailing from the UK and Portugal is set to sue the taxi-hailing company in an unprecedented ruling that could underscore the dangers of relying on performance monitoring tech

As businesses become automated and reliant on data and technology, it’s easy to see us drifting towards wider use of employee performance monitoring technology to manage our human workforces.

In an age of remote working especially, new tools are being peddled that enable employers to monitor staff activity, enabling a top-down, metric-based view of performance. Not just a virtual ‘peak over the shoulder’, these tools can generate data that, in some cases, can feed automated decision-making systems that can say, award a performance bonus in the form of a $50 voucher, or what about firing them on the spot?

For better or worse, performance monitoring and surveillance have the potential to put a rift in the power dynamics between workers and employers – a report by Clutch found that just 10% of workers would trust their employer’s use of performance tracking and monitoring software.

For years now, retail juggernaut Amazon has been notorious for using data and a stringent set of metrics to not only track the performance of workers in its warehouses but to automatically terminate their employment if they fail to meet certain standards.

In one facility in Baltimore, an investigation found that approximately 300 full-time workers were fired for inefficiency between 2017 and 2018 – a figure that represents 10% of the factory’s overall workforce.

‘Automated’ dismissals

In the first legal challenge of its kind, a group of former Uber drivers hailing from the UK and Portugal, are set to sue taxi-hailing company Uber for utilizing an algorithm to automatically fire them.

The ‘automated’ dismissals highlighted recently come as a test to the protections of GDPR Article 22. The regulation which came into force in 2018, imposes obligations on companies that collect people’s personal information and relates to all EU consumers regardless of where the company is located.

Considered as the biggest case on Article 22 of the GDPR to have reached the courts, the court ruling could lead to an unprecedented outcome following thousands of complaints from Uber drivers all over the world saying they had been dismissed automatically for committing fraud without an explanation.

The case has been backed by both the International Alliance of App-based Transport Workers (IAATW), and the App Drivers & Couriers Union (ADCU), a trade union for app-based workers in the gig economy.

“Uber has been allowed to violate employment law with impunity for years and now we are seeing a glimpse into an Orwellian world of work where workers have no rights and are managed by a machine. If Uber is not checked, this practice will become the norm for everyone,” said ACDU president Yaseen Aslam.

In response to the claims, however, Uber has said that the drivers’ accounts had been deactivated following a manual review by humans.

“Uber provides requested personal data and information that individuals are entitled to. We will give explanations when we cannot provide certain data, such as when it doesn’t exist or disclosing it would infringe on the rights of another person under GDPR. As part of our regular processes, the drivers, in this case, were only deactivated after manual reviews by our specialist team,” said an Uber spokesperson.

Power to challenge

The provisions provided by Article 22 have been debated for years, however, the right to an explanation about why an automated decision was made could give people more power to challenge if it is incorrect.

“This case demonstrates the impact of automated decision making for the millions of people that are working in the platform economy. For the first time, Uber drivers are challenging these decisions based on the GDPR,” said the four driver’s lawyer, Anton Ekker.

A study published in 2017 by Prof Edwards and Dr. Michael Veale of University College London, called Slave to the Algorithm? Why a ‘Right to an Explanation’ is probably not the Remedy you are Looking for, argues that the legal conception of explanations as “meaningful information about the logic of processing” may not be provided by the kind of ML “explanations” computer scientists have developed in response.

While enterprises are racing to benefit from new technologies, the latest news serves to underscore just how important it is to remember the dangers of relying on performance monitoring tech. Impersonal, data-driven decisions that impact workers’ daily lives can not only cause a significant loss of trust between employers and employees but can also serve to damage the reputation and integrity of organizations.