BRUSSELS, 27 October 2026 — European privacy campaigners have filed complaints against Scale AI in four countries, accusing the artificial intelligence company of illegally monitoring the freelance workers who label and check the data used to train AI systems. The complaints, reported by Reuters on Tuesday, 27 October 2026, target Scale AI's Outlier platform, where more than 900,000 freelancers around the world perform tasks that are recorded and turned into training data for clients including Meta, OpenAI and Google.
The case puts a spotlight on a workforce that is central to the AI boom but rarely seen: the people who rate answers, check facts, write examples and correct mistakes so that AI models learn to behave as their developers intend. The complaints argue that the way that work is watched, measured and rewarded breaks European privacy law — and they ask regulators to fine the company and to examine how it hands out work.
Illustration: A freelance data worker at a home desk, with abstract tracking symbols representing the monitoring described in the complaints.
What Happened
European digital privacy advocacy group noyb filed five complaints against Scale AI in France, Italy, the Netherlands and Austria on Tuesday, 27 October 2026, according to Reuters, which reviewed the complaints. The complaints focus on Outlier, Scale AI's platform for freelance data work, and allege that the company illegally monitored workers while they carried out tasks.
According to the complaints seen by Reuters, up to 19 trackers follow workers while they work, recording keystrokes and mouse movements every few seconds. An investigation by DiPLab, a Paris-based research collective, found that the resulting file could be rebuilt into a video that replays a work session from start to finish, Reuters reported.
The complaints also challenge the identity checks workers must pass. Workers must submit ID documents and a face video to identity verification firm Persona, and the complaints said those checks were invasive, repeated without explanation, and broke European Union rules on biometric data.
Scale AI did not immediately respond to a request for comment, Reuters reported. The complaints ask regulators to fine Scale AI and to examine how the platform assigns tasks.
Thomas Le Bonniec, who spent two years studying the platform for DiPLab, described a working arrangement in which flexibility runs in one direction. "You don't work when you want," he said. "You work when the platform wants you to and there are tasks available."
Background
Scale AI is one of the best-known companies in the business of supplying human-labelled data for AI development. Its clients have included some of the largest technology companies in the world, and Meta owns 49% of Scale AI, according to Reuters. The freelancers on Outlier do the detailed human work behind products that are often presented as purely machine-made: judging which answer is better, flagging unsafe content, and producing examples of good work for models to learn from.
That work is typically organised through a platform rather than an employment contract. Workers log in, pick up tasks, and are paid per task, often with bonuses for speed or volume. The complaints say Outlier also uses game-like mechanisms, and that bonuses are a key part of its incentive system. Outlier offered bonuses of $100 to $200 for completing 200 tasks in 24 hours, Le Bonniec said, citing interviews with workers during his investigation.
Italy’s data protection authority fined delivery platform Glovo in 2021 over similar gamification, Le Bonniec added, in a comparison that suggests regulators have already looked sceptically at reward systems that push gig workers to work faster and longer.
Under Europe's data protection rules, people generally have the right to obtain a copy of their personal data from an organisation that holds it. The complaints say that four workers asked Outlier for copies of their data between December 2025 and May 2026, and that none of them received an answer. That allegation, if upheld, would point to a failure of a basic transparency right, not just to an argument about how much monitoring is reasonable.
Illustration: A courthouse and a shield symbolising data protection law beside a phone running a gig-work app.
Why It Matters
The complaints matter for three reasons. First, they test where the line falls between legitimate quality control and unlawful surveillance. AI companies argue, in general, that they need to check human-labelled data carefully, because errors in training data can end up baked into widely used models. The complaints argue that recording keystrokes and mouse movements every few seconds, and reconstructing whole work sessions as video, goes far beyond checking the quality of a finished task.
Second, the case highlights the imbalance of power in platform work. Freelancers who depend on a platform for income may accept monitoring, identity checks and bonus-driven schedules that an employee in a conventional workplace could challenge more easily. When task allocation itself is under scrutiny — as noyb has asked regulators to examine — the question becomes whether workers are genuinely free to choose their hours, or whether the platform's design effectively controls them while keeping them outside employment protections.
Third, the outcome could ripple across the AI supply chain. If regulators in four countries find that this kind of monitoring breaks the rules, other data-labelling platforms operating in Europe would face pressure to change how they track workers, what identity evidence they demand, and how they answer data-access requests. Because the same platforms serve clients worldwide, changes made for Europe often influence practices elsewhere too.
The biometric angle adds a further layer. Face videos and ID documents are sensitive personal data, and European rules treat biometric data with particular caution. Repeated checks without a clear explanation, as alleged in the complaints, would be difficult to justify if regulators agree that less intrusive methods could achieve the same goal of verifying identity and preventing fraud.
What Happens Next
The complaints now sit with data protection regulators in France, Italy, the Netherlands and Austria. Each authority will decide whether to investigate, what evidence to seek from Scale AI, and whether the allegations warrant enforcement action. Regulatory cases of this kind typically take months rather than weeks, and the company will have an opportunity to respond to the allegations before any decision is made.
Observers will be watching several questions in particular. Will regulators treat session-reconstruction tracking as proportionate quality control or as excessive surveillance? Will they require changes to identity verification, such as fewer repeat checks or clearer explanations? And will they look more broadly at how tasks are assigned, as noyb has requested, rather than limiting the case to monitoring alone?
For the workers themselves, the most immediate practical question is transparency: whether they can obtain the data collected about them, understand how it is used, and challenge it if it is wrong. Whatever the regulators decide, the complaints have already made visible a workforce that the AI industry depends on but seldom discusses — the hundreds of thousands of people whose judgements, made task by task, shape the behaviour of the models now used by millions.
Illustration: Human hands sorting data fragments that flow into an AI system, representing the people behind AI training data.
Sponsored: explore today's featured sponsored offers
FAQ
Who filed the complaints against Scale AI?
European privacy advocacy group noyb filed five complaints in France, Italy, the Netherlands and Austria on 27 October 2026, according to Reuters.
What is Outlier?
Outlier is Scale AI's platform where more than 900,000 freelancers perform tasks that are recorded to create AI training data for clients such as Meta, OpenAI and Google.
What monitoring do the complaints allege?
The complaints say up to 19 trackers follow workers, recording keystrokes and mouse movements every few seconds, and that DiPLab found the resulting data could be rebuilt into a video replaying a work session.
What do the complaints say about identity checks?
They say workers must submit ID documents and a face video to verification firm Persona, and that the checks were invasive, repeated without explanation, and broke EU rules on biometric data.
What happens next?
Regulators in the four countries will decide whether to investigate and what action, if any, to take. Scale AI did not immediately respond to Reuters' request for comment, and no finding has yet been made.
Sources
Reuters, "Privacy group files complaints against Scale AI over alleged worker surveillance," 27 October 2026. Reuters reporting on the noyb complaints, the DiPLab investigation, and Scale AI's clients and ownership as cited in this article.
Comments
Post a Comment