Meta tightens glasses privacy safeguard amid reported sensing prototype
Meta says its Ray-Ban AI glasses will shut off the camera if the recording light is damaged, while a reported prototype raises new privacy questions.
By Hana Yoshida · Markets Reporter
4 min read
Meta is adding a new safeguard to its Ray-Ban Meta AI glasses that will disable the camera if the device detects that its recording light has been damaged or altered. The update addresses covert-recording concerns around smart glasses, even as the Financial Times reports Meta is testing a more sensor-heavy prototype.
In a blog post this week, Meta said the change applies to its second-generation smart glasses. The company said the glasses already turn off the camera when the LED indicator is covered, and the new measure extends that protection to cases where the light is tampered with or destroyed.
Meta said in the blog post that a blinking LED gives people nearby a visual cue that recording is underway. The company also said a camera-shutter sound loud enough for bystanders to hear would not be practical for the glasses.
Recording light draws scrutiny
Fortune reported that privacy concerns around smart glasses center on how easily they can capture images or video without drawing attention. A phone usually has to be held up and aimed, while glasses can record from the wearer’s point of view as the person looks around.
Meta has included the LED indicator on its smart glasses line since the first version launched in 2021, according to Fortune. Critics cited by Fortune have questioned the feature’s effectiveness because some bystanders may not understand what the blinking light means, may struggle to see it in daylight or may face users who have found ways to disable it.
Meta said in its blog post that it is taking down Facebook Marketplace listings from people offering to disable the LED indicator. The company said it may ban accounts or pursue legal action against people providing those services.
Dina El-Kassaby, a Meta spokesperson, told Fortune that users rely on the glasses for tasks such as music, live translation while traveling and hands-free calls. She said people wearing the glasses and people around them need to trust the product, and that Meta will keep strengthening protections as the glasses gain more capabilities.
Prototype report adds pressure
The Financial Times reported this week that Meta is testing prototype glasses with what it described as “super-sensing” abilities. According to the FT, the prototype would gather continuous audio and take photos every few seconds so users could later ask the glasses’ AI about what they had seen or heard.
The FT reported that Meta executives have discussed leaving the LED indicator off while those functions are active, though plans could change. The FT also reported that Meta would not store the raw footage and audio, and users would not be able to access that raw data.
El-Kassaby told Fortune that Meta does not comment on internal prototypes. She said the company is developing new technologies intended to help people during the day with privacy protections, and pointed to Meta’s Aria research glasses as one project designed to avoid capturing photos and videos in the same way traditional cameras do.
Fortune also reported that Meta was named earlier this year in a lawsuit alleging that private moments captured by users’ smart glasses were later viewed by workers in Kenya who were reviewing material to help train Meta AI models. The complaint, as cited by the Wall Street Journal, described footage of people inside homes during private activities.
Fortune reported that photos and videos captured by a user stay private unless the user chooses to share them. When media is shared with Meta AI, contractors may sometimes review it to improve the product, and Fortune reported that the material is filtered to protect privacy and remove identifying information.
Mark McCreary, a partner and chief artificial intelligence and information security officer at Fox Rothschild, told Fortune the anti-tampering update was a positive step. He also said the move appeared to clash with the FT’s report about a prototype that could collect more ambient information.
McCreary told Fortune that Meta’s advertising business adds to his concerns, saying more than 90% of the company’s revenue comes from advertising. He said AI glasses create a consent problem because wearers may agree to share material with Meta AI while bystanders captured in that material may not have agreed.
This story draws on original reporting from Fortune.