Uber Fined €824.99 Million Over Automated Driver Deactivations

Correction — September 25, 2026: An earlier version incorrectly described the €825 million penalty as a fine for transferring European driver data to the United States. That was a separate €290 million Dutch decision issued in 2024. The 2026 €824.99 million fine concerns automated decisions affecting drivers. This article has been rewritten to reflect the regulator's decision. The Dutch Data Protection Authority fined Uber €824,990,000 over automated individual decisions affecting drivers on its platform. The case matters because the disputed decisions were not minor recommendation-system tweaks. Regulators examined automated processes that could temporarily suspend or permanently deactivate a driver's access to the platform — a decision that can directly affect someone's ability to work. What the 2026 decision is actually about France's CNIL, which cooperated in the cross-border case, says the Dutch authority fined Uber for automated individual decisions concerning drivers. The underlying collective complaint represented more than 170 drivers and covered several issues, including transparency, international data transfers, and automated decisions. Those issues did not all become one fine. The 2026 decision focused on automated temporary and permanent disconnections from the platform. Reuters reported that the Dutch regulator objected to automated driver suspensions and deactivations without adequate information and meaningful human oversight. Uber disputes that characterization and has said it will appeal. Why automated deactivation deserves scrutiny When an algorithm controls access to work, three questions become operational rather than academic: Was the person told why the decision happened? Could a human meaningfully review and reverse it? Was the system making a decision with major effects based solely on automated processing? For a ride-hailing driver, an account suspension can mean lost income immediately. The fact that software can make that decision faster does not make the consequences smaller. The broader issue is algorithmic management: platforms increasingly use automated fraud detection, ratings, risk scoring, scheduling, and enforcement systems to govern workers. A system can be statistically useful and still create serious due-process and transparency problems when the affected person cannot understand or challenge its output. Do not confuse this with Uber's 2024 data-transfer fine The same complaint history also produced a separate Dutch enforcement action over transfers of European driver data outside the EU. According to the CNIL chronology: December 2023: €10 million over failures to properly inform drivers. July 2024: €290 million concerning transfers of data outside the European Union. August 2026: €824.99 million concerning automated individual decisions. That distinction matters. Combining separate cases makes it harder to understand what regulators actually objected to and what companies would need to change. Uber is appealing Uber contests the 2026 decision. Reuters reported that the company argues driver deactivations included human review and that the regulator's description overstates the role of fully automated decision-making. That dispute is important. A regulator's decision is an official enforcement action, but it is not the last possible procedural step when a company appeals. The durable takeaway is not that every automated moderation or fraud system violates European law. It is that systems making consequential decisions about people need defensible transparency, review, and governance. Why workers should care Automated management is becoming normal well beyond ride-hailing. Delivery platforms, warehouses, call centers, marketplaces, and conventional employers increasingly rely on software to flag fraud, score performance, schedule shifts, filter applicants, and restrict accounts. For workers, useful questions include: What decisions are automated? What data feeds the system? Does a human review adverse decisions before they take effect? Can the worker obtain a reason and challenge the result? Does an appeal pause the consequence, or does income stop first? Those are product-design questions, employment questions, and privacy questions at the same time. Sources CNIL — Automated decisions: UBER fined nearly EUR 825 million Reuters — Dutch regulator fines Uber for automating driver suspensions