Uber has been found to be in violation of the European Union's algorithmic transparency regulations following a legal challenge from two drivers whose accounts were terminated due to automated flags. The Amsterdam District Court ruled against Uber, which has accrued daily fines of €4,000 for its ongoing non-compliance, totaling more than €584,000.
The court ruled in favor of the two drivers seeking clarity on their data access, contesting what they describe as ‘robo-firings.’ However, the appeals court determined that Uber had sufficiently informed a third driver about the reasons behind the algorithmic flagging of their account for potential fraud. The drivers are pursuing Uber for information they believe they are entitled to regarding significant automated decisions affecting their accounts.
Under the European Union’s General Data Protection Regulation (GDPR), individuals have the right to not be subjected to solely automated decisions that significantly impact them, as well as the right to receive detailed explanations about such algorithmic processes. This includes “meaningful information” on the logic, significance, and potential consequences of the automated decision-making.
The core issue revolves around the automated flags that prompted fraud reviews of driver accounts, rather than the subsequent reviews conducted by Uber’s human staff. In April, a Dutch appeals court largely sided with platform workers against Uber and Ola regarding their data access rights tied to alleged robo-firing. The ruling stipulated that these platforms could not invoke trade secrets to deny drivers access to their data related to AI-powered decisions.
In this recent case, Uber attempted to reassert its argument around commercial secrets to limit data disclosure to drivers regarding the reasons for its algorithmic flags. The company also contends that revealing operational details would compromise its anti-fraud systems.
In the instances of the two drivers who succeeded in their claims, the court found Uber had failed to provide any information about the automated flags that led to their account reviews. This result acknowledged an ongoing breach of the EU's algorithmic transparency regulations. The judge speculated that Uber might be deliberately withholding information to protect its business model and revenue streams.
In contrast, for the third driver, the court concluded that Uber had provided “clear and, for the time being, sufficient information.” The ruling indicated that the automated flagging process for this driver relied on several criteria, including the number of canceled rides, the total rides completed, and the ratio of cancellations to completed rides. The court noted that due to their unusually high cancellation fee rate, the driver was flagged for potential cancellation fee fraud.
The driver argued for more clarity in Uber’s explanation, claiming the information offered was still vague and lacked meaning, as he did not understand the thresholds Uber uses to label a driver as fraudulent. Nevertheless, the interim relief judge sided with Uber, stating that providing such details could make it too simple for others to exploit the system without consequences.
As of now, the broader implications of whether Uber correctly categorized these drivers as fraudsters remain unaddressed in the litigation process. The ongoing legal battle aims to clarify the extent of information that platforms using algorithmic management must share with workers upon request, as per EU data protection laws, while also recognizing the complexities of protecting AI operational details from exploitation.
Following the ruling, an Uber spokesperson stated, "This ruling pertained to three drivers who lost access to their accounts under very specific circumstances. When these accounts were flagged, they underwent review by our Trust and Safety Teams, specially trained to identify behaviors that could jeopardize rider safety. The Court affirmed that these reviews were standard practice following our system’s identification of potentially fraudulent activities."
The drivers are supported in their legal fight by the advocacy group Worker Info Exchange (WIE) and the App Drivers & Couriers Union. Anton Ekker, their legal representative, stated, “Drivers have been advocating for their right to information on automated deactivations for years. The Amsterdam Court of Appeal affirmed this right in its pivotal April 2023 judgment. It is deeply troubling that Uber continues to ignore the Court’s directive, but I am confident that the principle of transparency will ultimately succeed.”
James Farrar, director of WIE, echoed this sentiment, remarking, “Whether it’s the UK Supreme Court for worker rights or the Netherlands Court of Appeal for data protection, Uber consistently ignores the law, defying even the highest court orders. Drivers have endured years of relentless algorithmic exploitation while fighting for justice, as governments and regulators fail to enforce existing regulations. Meanwhile, the UK government appears to be dismantling the minimal protections available to workers against automated decision-making through the Data Protection and Digital Information Bill currently under discussion in Parliament.” The proposed EU Platform Work Directive similarly risks becoming ineffective unless serious enforcement measures are established.
Ranking
