German rights group lodges criminal complaint over Meta smart glasses privacy
The nonprofit targets Meta, Ray-Ban, Oakley and retailers, calling for dashcam-style recording restrictions amid concerns over bystander privacy and third-party AI data processing.

German human rights organisation HateAid has filed a criminal complaint against Meta, Ray-Ban, Oakley and several retailers, alleging that the sale of Meta’s smart glasses violates German privacy legislation. The complaint contends that the devices’ covert recording capabilities constitute a criminal offence under the Telecommunications, Digital Services and Data Protection Act.
HateAid is urging regulators to impose recording restrictions comparable to those applied to dashcams and bodycams. The group argues that selling devices capable of discreetly capturing video and audio on public streets poses an unacceptable risk to bystander privacy, particularly given that individuals in public spaces have not consented to Meta’s terms of service.
In response to the allegations, a Meta spokesperson told Politico EU that the smart glasses were built with privacy safeguards from the ground up that go beyond what any smartphone offers. The company states that the devices feature an LED indicator that flashes while recording, alongside software designed to block camera use if tampering with the light is detected.
However, privacy advocates point out that these safeguards can be circumvented. Several companies are now selling aftermarket accessories designed to block the recording LED without triggering Meta’s tamper-detection software, effectively allowing users to record covertly. This has raised concerns that the primary function of the privacy settings is to protect the wearer rather than the subjects being recorded.
Broader concerns have also emerged regarding how data is processed. While standard photos remain on the device unless shared, AI features send information about surroundings to Meta for processing. Third-party contractors are employed to review and label some of these interactions for artificial intelligence training.
The risks associated with this data pipeline were highlighted earlier this year when a Swedish newspaper reported that contractors in Kenya encountered intimate and sensitive material captured by the devices. The reported content included nudity, bathroom footage, sexual acts and credit card numbers.
This complaint follows other recent scrutiny of the product’s capabilities. In June, Meta removed dormant facial-recognition code from its smart glasses companion app after it was discovered by Wired. The legal outcome of HateAid’s criminal complaint remains uncertain.

