Meta working on a new AI transcribing pendant? Here's what we know so far

Apart from its range of smart glasses, Meta also appears to be planning a new AI-powered pendant capable of recording and summarising conversations throughout the day. According to a report from The Information, Meta acquired Limitless in December 2025.

The startup had developed a device called the Limitless Pendant, a clip-on Bluetooth microphone that, for a starting price of $199, offered users roughly 20 hours of recording, transcription, and summarisation every month.

Using the device, users could attend in-person meetings and later access AI-generated transcripts and meeting notes through the Rewind app. Production of the pendant has since ceased following Meta's acquisition of the company, although Meta has promised to continue supporting existing users for a year after the acquisition.

An internal memo from Alex Himel, Meta's Vice President of Wearables, reportedly also revealed plans to expand the company's current lineup of smart glasses and launch a subscription service aimed at businesses called "Wearables for Work."

More from Tech

The report comes as Meta pushes ahead with ambitious plans to expand its portfolio of AI-powered wearables, despite continued losses at its hardware division. While products such as the Meta Ray-Ban smart glasses have reportedly performed well, Meta's Reality Labs unit posted a loss of $4.03 billion in the first quarter of 2026 on revenue of just $402 million. According to The Information, the company is targeting sales of 10 million wearable devices during the second half of 2026.

However, always-listening AI wearables have faced growing resistance in the real world. Friend, an AI-powered necklace designed to listen to users' surroundings and offer companionship, saw its promotional posters vandalised in New York shortly after launch.

At the same time, privacy concerns surrounding devices such as the Meta Ray-Ban glasses are attracting increasing scrutiny. Speaking to BBC News earlier this month, David Harris, a former Meta AI researcher and lecturer at UC Berkeley, said that technology "like this is fundamentally an invasion of privacy" and is likely to face growing public backlash in the years ahead.