Read the technical blog here: [Link | Archive]
Meta has fully developed a working facial recognition pipeline inside its Ray-Ban smart glasses companion app that can trigger a notification naming an identified person when a known face is detected through the glasses’ AI-powered cameras, according to code inspection conducted by Buchodi. Buchodi is a security-minded engineer who audits the security of software development kits (SDKs).
Developed, but not implemented: However, Meta has not made the feature operational yet. As a result, regular users of Meta Ray-Ban smart glasses cannot currently use it. However, the functionality appears ready to be deployed if Meta chooses to enable it.
How does it work? The researcher states that the code contains three face-recognition models, totalling around 100 MB in size, with the following functions:
- android_facerec_scrfd (3.4 MB): Detects faces in an image.
- android_facerec_kps_aligner (117 KB): Crops and aligns each detected face.
- android_facerec_sface (96 MB): Converts a face into a 2,048-dimensional embedding (a biometric fingerprint).
The researcher further pointed to the following open-source architectures on GitHub as the technology underpinning the system:
- InsightFace: A state-of-the-art 2D and 3D face analysis project. [ GitHub | Website ]
- SFace: “SFace can make a better balance between decreasing the intra-class distances for clean examples and preventing overfitting to the label noise, and contributes more robust deep face recognition models.” [ GitHub ]
Code is not the evidence of privacy violation: “Worth stating plainly: shipping detection and embedding models is not, by itself, evidence of recognition. Plenty of apps run on-device face detection for framing or autofocus.”
What can it do if implemented in its current form? According to the researcher, “Run end-to-end, it recognizes a known face and names it in a notification, and it stages unknown faces (crop + fingerprint) to disk.”
“What I am not claiming: that Meta is identifying strangers for users today, that enrollment data is flowing, or that any of this is enabled in production,” the researcher added.
Architecture at this level can’t be an accident: “What’s hard to wave away: building, shipping, and wiring this much apparatus down to a 2048-dimension facial fingerprinting and a hardcoded ‘Person recognized’ notification, is an engineering investment. Capability that doesn’t ship by accident,” the researcher concluded.
Meta has access to users’ sensitive and private data, including nudity: In March 2026, an investigation by Swedish newspapers revealed that Meta allowed data annotators in Nairobi, Kenya, working for Sama, a San Francisco-based contractor, to access sensitive, unanonymized data from Meta Ray-Ban smart glasses users. The investigation also found that these workers had access to live data streams. Swedish…
Source link
Disclaimer
We strive to uphold the highest ethical standards in all of our reporting and coverage. We blogs.grocliq.com want to be transparent with our readers about any potential conflicts of interest that may arise in our work. It’s possible that some of the investors we feature may have connections to other businesses, including competitors or companies we write about. However, we want to assure our readers that this will not have any impact on the integrity or impartiality of our reporting. We are committed to delivering accurate, unbiased news and information to our audience, and we will continue to uphold our ethics and principles in all of our work. Thank you for your trust and support.
Website Upgradation is going on for any glitch kindly connect at [email protected]
