Meta knows you probably have some privacy concerns about its smart glasses.
At its annual Meta Connect event today in Menlo Park, California, the company announced plans to incorporate its siloed AI processing scheme known as Private Processing into its Ray-Ban Meta smart glasses. The feature, designed for WhatsApp last year, allows the glasses’ AI assistant to get to know you and fulfill your requests without Meta accessing any of your data.
“Most of the time, glasses are helping you see well, protecting your eyes and complementing your look, and that’s it,” Meta wrote in its press release, perhaps somewhat underselling the monumental historic significance of vision-correcting lenses as a technology. “But AI glasses have the capacity to provide superpowers like translating a conversation or summarizing notes or a conversation.” So the company says it sought to answer the question: “How do you build a hyper-personalized AI that knows your world deeply with enhanced privacy?”
Private Processing is designed to enable AI cloud processing that protects data from all outside parties, including Meta itself. For its smart glasses, Private Processing could theoretically enable chat interactions wearers have with Meta’s AI assistant through its smart glasses and commands, including private transcriptions and messages. However, the company has not yet said specifically which features it will support.
Meta introduced Private Processing into WhatsApp and its Meta AI app last year. The service helped enable WhatsApp’s Incognito Chat. The goal was to let customers use AI services without undermining WhatsApp’s flagship end-to-end encrypted secure messaging feature.
In a short conversation at Meta Connect, WIRED spoke with Meta’s Pritam Shah, director of Engineering, on adding Private Processing to the glasses.
“What we are doing right now is sharing our excitement about this enhanced privacy platform,” Shah says. “We have the platform ready. We have it working. It’s not just a white paper and a blog. The reason why we are announcing it right now before putting more experiences on this platform is so that we can start working with the security community and the privacy community and a third-party auditor.”
All of this comes at a moment of reckoning for Meta. The company has long been criticized for how it handles user privacy. This year alone, Meta has fielded months of controversy from people worried about its use of facial recognition technology and how its smart glasses could be used to arm sexual predators. Some people have dubbed them “pervert glasses,” confronted people wearing them, and even built apps like ZuckOff to detect Meta’s smart glasses nearby.
More broadly, Meta has been dogged by years of security and privacy gaffes that have eroded user trust, and its products have often seemed to be better at surveilling users than helping them. But at Meta Connect, the company tried to position its smart glasses as a tool that isn’t creepy (it also announced several new models, including a pair of VR glasses).
When the company launched its Muse AI agent earlier this month, it positioned the assistant as being geared toward security and privacy, specifically touting Muse’s cloud environment, known as Secure VM. It’s working on a version, dubbed Confidential VM, that will feature enhanced privacy guarantees, including a similar promise to make data inaccessible to Meta itself. Muse will also be coming to Meta’s glasses.
Private Processing isn’t available yet on Meta’s glasses, and Shah doesn’t have a timeline for when it will be added or which specific features will work with it.
“These are examples,” he says. “These are not concrete features coming, but these are some examples. And there will be even more powerful examples.”