Uber has been hit with a €825 million ($966 million) fine in the Netherlands over allegations that automated systems suspended or deactivated drivers without sufficient human oversight, creating one of Europe’s biggest tests yet of how platforms can use algorithms to make high-impact decisions about workers.
TechCrunch reports that the Dutch Data Protection Authority imposed the €825 million penalty, making it the second-largest fine issued so far under the European Union’s General Data Protection Regulation. The regulator investigated complaints alleging that Uber deactivated driver accounts through automated processes without adequate warning or human intervention.
Anadolu Agency similarly reports that the authority found Uber used automated systems to deactivate drivers’ accounts without properly informing them or giving them adequate opportunities to challenge the decisions. The case concerns conduct between 2020 and 2022 and originated from complaints by drivers in France.
Regulators Are Drawing a Line Around Automated Decisions
The ruling targets more than Uber’s fraud-detection systems. It addresses where automation should stop when software decisions can materially affect someone’s income.
Dutch regulator deputy chair Monique Verdier said a computer should not make decisions alone when the consequences are so significant, according to TechCrunch. The regulator concluded that Uber had committed “serious infringements” in its treatment of drivers.
Anadolu reports that EU data-protection rules prohibit algorithms from independently making decisions that have a significant impact on people’s lives without appropriate human oversight, while affected people must also have an opportunity to challenge those decisions.
For product and engineering teams building marketplaces, fintech systems or workforce platforms, that makes human review an architectural requirement when automated decisions cross certain risk thresholds.
Uber Disputes the Regulator’s Findings
Uber argues that its processes already include human involvement.
TechCrunch reports that the company said most suspensions are brief, no permanent deactivation occurs without human review, and drivers can appeal decisions. Dutch regulators, however, said some drivers were permanently deactivated without human review—a conclusion Uber disputes. The company plans to appeal.
Anadolu likewise reports that Uber “fundamentally disagree[s] with this decision” and considers the penalty disproportionate. Uber also said people are now involved in deactivation decisions and drivers can challenge them.
A Driver Campaign Became a Platform-Governance Case
The dispute traces back to drivers challenging h1918ow Uber’s systems classified suspected misconduct.
Anadolu reports that some drivers were temporarily suspended over suspected fraud, including allegations of taking unnecessary detours to increase earnings. The case was handled in the Netherlands because Uber’s European headquarters are in Amsterdam.
TechCrunch reports that former French Uber driver Brahim Ben Ali, whose account was deactivated in 2019, collected testimony from 170 other Uber drivers before pursuing the complaint with assistance from Swiss digital-rights nonprofit PersonalData.io.
Founder Paul-Olivier Dehaye told TechCrunch that the same group of drivers was behind complaints that previously contributed to separate Uber penalties of €290 million and €10 million.
Human Oversight Is Becoming Part of Product Infrastructure
For technology companies, the wider lesson is not that automated fraud detection is disappearing. Platforms handling millions of users still need software to flag anomalous behavior at scale.
The harder question is what happens after the algorithm flags someone.
Uber’s case shows that regulators increasingly care about whether automated systems produce an accusation, a temporary restriction or the final decision itself—and whether a person can understand, challenge and reverse that outcome.
For builders of marketplaces and AI-driven operational systems, that turns appeals, explainability and human escalation from secondary support functions into part of the core product stack.
