The latest issue of our fashion and luxury newsletter covers recent antitrust, data privacy and artificial intelligence developments.
Ray-Ban Meta AI Glasses under scrutiny: Key findings from the Hamburg Data Protection Commissioner’s Final Report
On 10 September 2026, the Hamburg Commissioner for Data Protection and Freedom of Information (the Authority) published its final report on the technical and data protection review of the Ray-Ban Meta AI Glasses (Gen 1 model, introduced in 2023). The report presents the results of a comprehensive investigation covering the device's hardware, software, data flows, and the legal framework governing its use under data protection law.
A notable finding was the discovery of database tables within the Meta AI app whose names suggest facial recognition functionality (including tables named “face”, “face_group”, “face_low_confidence_pair” and “face_to_face_group”). While the tables were found to be empty, the Authority concluded that the basic software structures for implementing facial recognition are already in place, even if no such feature is currently active.
The report also examined the effectiveness of the LED recording indicator on the glasses – intended to notify people being recorded – and found significant limitations. For example, during video recording, the system checks whether the LED is covered only at the moment recording begins. If the LED is covered after recording has started, the recording continues without any visible indicator, rendering the transparency mechanism completely ineffective.
From a data protection perspective, the Authority assessed the key processing activities in three areas: photography and video recording, the use of Meta AI, and the publication of recordings on Facebook and Instagram. The legal analysis focuses on the responsibilities of glasses users as data controllers within the meaning of the General Data Protection Regulation (GDPR) with respect to the personal data of third parties.
On the question of AI training, the report highlights that Meta uses an opt-out model, meaning that interaction data – including, potentially, photos of third parties – is used for the development of generative AI models unless users actively object. The Authority concluded that when AI training is enabled, the so-called “household exemption”1 under the GDPR does not apply. This is because the processing extends beyond private purposes and involves the transfer of personal data into a training context determined by Meta. In this scenario, users become joint controllers with Meta under Article 26 GDPR with respect to the collection and transfer of training data.
The report further concludes that the use of third-party personal data collected via Ray-Ban Meta AI Glasses for AI training purposes generally cannot be based on either valid consent or legitimate interests. Consent typically fails because third parties are unaware of the processing and cannot be informed in real time. The legitimate interests test under Article 6(1)(f) GDPR also generally fails: uninvolved third parties are included in a global AI training context without sufficient transparency, effective control and reasonable foreseeability.
The Authority’s report has important implications for the fashion and luxury eyewear sector, particularly for companies involved in the development, manufacture or distribution of wearable AI-enabled devices. The presence of facial recognition database structures, even if not yet activated, raises fundamental questions about the direction of future product development and the risk of scope creep. Companies in this space should ensure compliance with GDPR requirements, including robust transparency mechanisms, effective consent procedures, and careful assessment of AI training practices involving third-party data.