Meta AI glasses may soon watch and listen to you all day, report says

Meta could be working on one of its most ambitious AI projects yet. According to a report by the Financial Times, the company is testing a new generation of AI-powered smart glasses that may constantly observe a user's surroundings, allowing its AI assistant to remember conversations, recognise places, and answer questions about what happened during the day.

The reported feature, internally referred to as "super sensing", would be from today's AI glasses, which respond only after users activate them. Instead, the new system is said to remain aware of the environment for extended periods, collecting visual and audio information in the background so Meta AI has more context when responding to queries.

If introduced, the technology could make AI glasses significantly more capable. Rather than asking the assistant to identify something in front of them at a particular moment, users may be able to ask questions later about people they met, objects they saw, or conversations they had earlier in the day.

Always-aware AI, but with privacy questions

The report says the prototype glasses are designed to capture photos every few seconds while also listening to surrounding audio. However, Meta is reportedly exploring a system that does not permanently store the raw recordings. Instead, the company may extract metadata from the captured information and upload only that data to its servers, where Meta AI can process it to answer user requests. The idea, according to the report, is to reduce the privacy risks associated with storing continuous recordings.

Even so, the concept is likely to raise fresh concerns around surveillance and consent. Meta's smart glasses have already attracted criticism over privacy, with some people questioning whether bystanders can always tell when they are being recorded. The company has also faced scrutiny over reported work involving facial recognition and attempts by some users to disable the LED recording light that indicates when the camera is active.

Meta recently announced a software update that disables the camera if the glasses detect tampering with the LED indicator. However, the Financial Times claims that the company is considering keeping the indicator turned off while the reported "super sensing" feature is running because it would be treated as an AI function rather than an active photo or video recording mode.

The report further states that Meta is evaluating whether information gathered through the feature could also be used to improve its AI models. It also suggests that the capability may eventually arrive on some existing Meta smart glasses through software updates instead of being limited to future hardware.

Meta's long-term vision for AI glasses

Responding to the report, Meta spokesperson Dave Arnold told The Verge, "While we don't comment on internal prototypes, we're committed to getting our glasses right because they need to be loved by both people wearing them and those around them." He added, "Our approach has been to develop new technologies that will help people throughout their day, with privacy built in from the ground up."

The reported development also fits into Meta's vision for AI wearables. CEO Mark Zuckerberg has previously said he wants smart glasses to evolve into an AI assistant that stays with users throughout the day, helping them remember information and complete everyday tasks. In an earlier earnings call, Zuckerberg said he was excited about the possibility of glasses becoming "a personal agent" that remains with users all day instead of simply responding to individual questions.

Meta has also indicated in previous announcements that it wants its AI glasses to become more proactive over time, offering assistance without requiring users to repeatedly trigger the assistant.