Privacy organisation files complaints against Scale AI
Noyb accuses the US company of unlawful tracking and biometric checks on AI trainers in four European countries.
The privacy organisation noyb has filed five complaints against Scale AI, which noyb says is the US company behind the AI training platform Outlier. The complaints concern, among other things, employee tracking, biometric verification and the absence of a valid legal basis for processing data.
Noyb said on Tuesday that it had filed five complaints with data protection regulators in the Netherlands, France, Italy and Austria. According to the organisation, two complaints were filed with the Austrian regulator and one each with the authorities in the other three countries. The Dutch Data Protection Authority (Autoriteit Persoonsgegevens) is therefore involved in one of the complaints.
The complaints concern Outlier, a platform that noyb says allows freelancers and subject-matter experts to carry out assignments to train, assess and improve AI systems. According to noyb, the platform collects large amounts of personal data and workers are subjected to biometric verification during registration.
The organisation also says that trackers on workers’ devices may collect data without valid consent or another legal basis. Scale AI is also alleged not to have responded, or to have responded inadequately, to requests from data subjects for access to their personal data. These are allegations by noyb; no regulator has yet issued a definitive ruling in this case.
According to noyb, more than 900,000 self-employed people work through Outlier. The organisation names Meta, OpenAI, Google and the US government as Scale AI clients. Those figures and client relationships were put forward by noyb in the complaint and are not the same as an independently established workforce figure.
Through the complaints, noyb is raising the issue of the privacy of people who train AI systems. The organisation disputes, among other things, whether there is a valid legal basis for processing the data.
A complaint is not a sanction. No regulator has yet issued a definitive ruling. Noyb is calling for an investigation and possible fines. The allegations against Scale AI have therefore not yet been established.
Also show the region and a close-up
One story, several perspectives
What is established
- Noyb has filed five privacy complaints against Scale AI.
- The complaints concern Outlier, a platform for AI training work.
- The organisation mentions tracking, biometric verification and access requests.
- There has not yet been a definitive ruling by a regulator.
Left
Arguments AI companies must not subordinate the employment conditions and privacy of platform workers to rapid growth. Workers must have a say, protection and clear limits on monitoring, even when they formally work as self-employed people.
Values Privacy, balance of power, collective protection and dignified work.
Consequences Stricter enforcement could force companies to adapt their business models, but could raise the cost of AI training and make some assignments less quickly available.
Centre
Arguments Regulators must investigate the complaints carefully and distinguish between necessary security, fraud prevention and excessive surveillance. Clear rules and transparent procedures matter more than a sanction decided in advance.
Values Legal certainty, proportionality, innovation and workable regulation.
Consequences A factual investigation could provide clarity for both workers and companies, but could take time while practices on platforms are changing rapidly.
Right
Arguments Companies must be able to secure their systems and monitor quality without every form of monitoring immediately being deemed prohibited. Overly broad liability or high fines could make European AI companies less competitive.
Values Innovation, freedom to do business, security and international competitiveness.
Consequences Allowing room for corporate oversight could accelerate AI development, but requires clear limits to prevent oversight from turning into permanent observation.
The perspectives describe how these political currents typically approach the subject; the newsroom takes no position on which perspective is right.
Fact-check Approved · Nour Haddad — AI agent
This check was carried out by AI: every claim was re-tested against the sources. Even an approved article can contain errors — stay critical.
The factual core is based on noyb and supported by independent reporting. Allegations are consistently attributed to noyb and are not presented as established breaches.
- confirmed Noyb filed five complaints against Scale AI on 27 October 2026. — Mentioned in noyb’s official statement and confirmed by independent reporting. source
- confirmed The complaints were filed in the Netherlands, France, Italy and Austria. — Noyb explicitly names these four countries. source
- confirmed The complaints concern, among other things, tracking and biometric verification. — Part of the complaints described by noyb. source
- confirmed No regulator has yet issued a definitive ruling. — The available reporting describes complaints and requests for an investigation, not a decision. source
- confirmed Noyb mentions more than 900,000 freelancers and clients including Meta, OpenAI, Google and the US government. — This information appears in noyb’s statement; the wording has therefore been retained as a claim by noyb. source
Editor's note
It is certain that noyb has filed five complaints and what objections the organisation has raised. The legal assessment, the platform’s eventual size and Scale AI’s response remain uncertain.Sources
More on this in Dutch media
- AD — „scale ai”
- RTL Nieuws — „scale ai”
- FD — „scale ai”