In a recent article entitled “Shades of Glory” published in the New York Times Sunday magazine, the author said of Meta AI glasses, “[T]hey are basically a whole sting operation that sits on your nose.”
[1] Truer words were never spoken.
Meta AI glasses convert the user’s eyeglasses to an in-lens smartphone. The glasses come in a range of models, all of which feature a camera to take videos and images, a microphone to capture audio, and an AI-based “assistant” to interact with the user in various ways. The AI assistant provides responses to oral questions and prompts, translates languages, and sends messages and phone calls. The glasses also enable wearers to direct the built-in assistant to take photos, videos, and livestream directly to Meta’s social media platforms, Facebook and Instagram. Meta’s AI assistant is activated when the user says “Hey Meta.”
Meta AI glasses are the product of a joint venture between Meta and Luxottica of America. Meta owns and operates some of the world’s largest social media and communications platforms, including Facebook, Instagram and WhatsApp and Threads. Luxottica of America controls a significant portion of the eyewear market in the United States through its ownership of brands including Ray-Ban, Oakley, Wayfarer and Sunglass Hut and its operation of major retailers such as LensCrafters, Pearle Vision, and Target Optical. Together Meta and Luxottica designed, developed, market and sell Meta AI glasses. By the end of 2025, over seven million pairs of Meta AI Smart Glasses had been sold.
Meta AI glasses are marketed with an emphasis on a commitment to privacy. Among the representations made are that the Meta AI glasses are “designed for privacy, controlled by you,” that the glasses are “built for your privacy and others too” and that steps are taken to “protect people’s privacy”.
Notwithstanding the representations, in February 2026, Swedish newspapers Svenska Dagbladet and Göteborgs-Posten released an investigative report revealing that contract workers at the Nairobi based data annotation company Sama review audiovisual recordings taken by the Meta AI Glasses. The footage captured by Meta’s AI glasses included audiovisual recordings of highly private and even intimate conduct, including naked bodies, sexual activity, and private bodily functions such as users in the bathroom. For example, data annotators reported reviewing footage of the users’ spouses changing in their own bedrooms after the glasses had been placed on a bedside table. Data annotators also reported reviewing footage that revealed confidential information, including images of financial statements and personal text messages.
Users cannot opt out of having their data processed by Meta’s servers when they use the AI assistant feature, and thus cannot avoid having their data transmitted to and reviewed by Sama employees, unless they want to use the glasses solely as standard eyewear, which defeats the purpose of the glasses.
This Swedish article spawned worldwide outcries. In the United States, several class actions were filed on behalf of a class of Meta AI glass owners alleging violations of privacy laws and false advertising. These lawsuits seek injunctive relief compelling better disclosure of the true nature of the sharing of images and monetary damage for diminished value of the eyeglasses due to the lack of privacy. In the United Kingdom, its Information Commissioner’s Office wrote to Meta to demand answers about how the company was meeting its obligations under UK data protection laws. Other countries in the European Union appear to be following suit with their own investigations.
And, how about all the unsuspecting people that are unwittingly filmed by users wearing the glasses. And perhaps even worse, how about every receptionist, cashier and security worker who can photograph sensitive information surreptitiously and use it for identity theft. The small green light on the eyeglasses’ frame is hardly visible enough to every person going through their busy day to alert them to the fact that they, or their personal documents, are being photographed. Beyond that, there are dozens of videos on YouTube describing how to disable the light and accessories are sold to block it.
In 2024, two Harvard students built a system that connected the Meta AI glasses to a facial recognition program, thereby demonstrating how the glasses could identify strangers in real time, pulling up names and family members within minutes. In response, Meta said “…our terms of service clearly state that users are responsible for complying with all applicable laws and for using Ray-Ban Meta glasses in a safe, respectful manner.” If stopping bad or illegal behavior were as easy as including a term of service prohibition, the world would be a safer place.
There are numerous reported incidents of people being filmed with Meta AI glasses without their consent. In October 2025, a man in San Francisco was reported to have used Meta AI glasses covertly to record women at the University of San Francisco. The university’s public safety officials sent an alert to students saying the man “has approached women with unwanted comments and inappropriate dating questions.” He was believed to have shared the encounters on his TikTok and Instagram pages.
The BBC reportedly has seen hundreds of similar short videos on TikTok and Instagram posted by dozens of male influencers claiming to offer advice on how to pick up women, which videos appeared to have been filmed secretly using Meta AI glasses. For example, a woman in London was approached by a man who started a conversation with a pickup line, followed her onto the elevator while continuing to chat, and asked her for her phone number. The woman did not realize the man secretly was filming her on his smart glasses; he subsequently posted the footage to TikTok, where it received 1.3 million views. He also had captured her phone number, resulting in the woman receiving an unwanted stream of messages and calls.
In December 2025, a Florida woman shared a video of herself that had been secretly recorded by a man who approached her at Tampa airport, filmed the conversation and posted it on Instagram. She had no idea he had recorded the whole conversation with his glasses, which she described as “creepy.”
In another incident in the UK reported by the BBC, a woman was filmed on a beach by yet another man wearing sunglasses. Unaware that she was being filmed, the woman chatted with the man, and during the conversation, she shared details about her employer and family. Representing them as dating advice, the man posted two videos online which garnered 6.9 views on TikTok and more than 100,000 likes on Instagram.
In January 2026, a woman in Vancouver was filmed with smart glasses in a fast-food restaurant. A man walked up to her wearing sunglasses, asked her for her name and told her she was gorgeous. She tried to reject him by saying she had a boyfriend. Imagine her surprise when her friend found the video among more than 100 similar videos on the man’s social media channels.
What is particularly worrisome is that these women were unaware they were being filmed for distribution, so they revealed snippets of information about their private lives; for example, a name, employer, that they live in the area. And, beyond personal information disclosed in conversation, the contractors at Sama reported seeing credit card numbers and other personal identifiers.
In a possible indication of how the public feels about the use of Meta AI glasses, in December 2025, a New York subway rider accused a woman of breaking his Meta AI glasses. The aggrieved glass wearer posted a video of the woman with the message “she just broke my Meta glasses.” The video garnered millions of views. Instead of sharing the man’s outrage, the internet celebrated the woman as a folk hero, with a plethora of comments along the lines of “good people are tired of being filmed by strangers.” Notwithstanding the pushback, it appears companies are making a strenuous effort to promote ever-increasing sales. It remains to be seen whether any of these legitimate privacy concerns will be addressed.
*******************
The author’s law firm is counsel for certain plaintiffs in privacy related litigation over Meta AI glasses.
[1] Sam Anderson,
Shades of Glory, New York Times Magazine, April 19, 2026.