Meta has reportedly discussed introducing facial recognition to its smart glasses, allowing users to identify people they come across. The technology would be introduced as part of a feature internally referred to by the U.S. company as “super sensing.” This feature keeps the smart glasses’ cameras and sensors on and uses artificial intelligence to help the wearer remember what they encountered during the day. For instance, the glasses could remind the user to take their keys when going out. Currently, Meta is testing an early version of super sensing, which could become a part of its “live AI” offering, according to a report from The Information.
Last year, Meta introduced an always-on AI assistant into its camera-equipped Ray-Bans, which can be activated by the prompt “Hey Meta, start live AI.” This AI assistant allows the wearer to use their voice to search and play content from platforms such as Spotify. Additionally, it supports automated translation to four languages, enhancing the usability for a diverse user base.
The glasses, code-named Aperol and Bellini, are expected to be released next year, in 2026. As part of their development, Meta is also trying to extend the battery life of its smart glasses, aiming to allow them to run AI features for hours instead of the current 30 minutes.
Meta’s interest in integrating facial recognition into its wearables comes after the company updated privacy policies for its smart glasses in April 2025. The updated policy specifies that “Meta AI with camera use is always enabled” unless the user manually turns it off. Furthermore, Meta has removed the option to disable storing voice recordings, raising concerns about user privacy.
As the company positions itself to potentially launch these advanced smart glasses, the broader context of facial recognition technology in consumer devices is noteworthy. Facial recognition through smart glasses has so far been limited mainly to use by law enforcement agencies. Countries such as China, Russia, and the United Arab Emirates have tested facial recognition-enabled smart glasses. In the U.S., Clearview AI signed a contract with the U.S. Air Force in 2022 for supplying facial recognition smart glasses.
Clearview AI, which does not sell its technology to the public, had previously hinted at plans to integrate its facial recognition technology into augmented reality glasses made by Vuzix, a U.S. company. This interest in facial recognition technology comes amid a backdrop of hesitation from consumer tech giants like Google and Facebook, particularly following a 2015 lawsuit against Facebook’s use of facial recognition to tag friends in photos.
However, the landscape is changing. Companies such as Clearview AI and PimEyes have made facial recognition technology more ubiquitous and accessible. In a striking demonstration of the technology's implications, two Harvard students made headlines last year after converting Meta’s smart glasses into a device that automatically captures people’s faces with facial recognition and runs them through face search engines, including those belonging to PimEyes. The software, named I-XRAY, streamed video from the glasses, capturing faces that were then matched against pictures on the internet. It scoured various data sources to find names, phone numbers, home addresses, and names of relatives of the recorded individuals. The students, AnhPhu Nguyen and Caine Ardayfio, explained that the main goal behind their project was to highlight the privacy risks presented by widely available technology.
In a different sector, the National Institute of Standards and Technology (NIST) has issued a draft update to its Privacy Framework to incorporate the latest cybersecurity guidelines and practices. The NIST Privacy Framework: A Tool for Improving Privacy through Enterprise Risk Management, was first released in January 2020. This framework is a voluntary tool that provides a set of strategies for organizations to adopt to improve their approach to protecting personal data.
The update comes five years after the initial release and aims to enhance usability, address current privacy risks, and maintain alignment with the recently updated NIST Cybersecurity Framework (CSF). The Core activities of the NIST Privacy Framework version 1.1 align with the Core activities of the NIST CSF 2.0, which was released in February 2024. Notable changes in version 1.1 include updates to cover privacy risks associated with AI and chatbot tools that were not widely available when the Privacy Framework was first introduced.
In addition to the structural updates, NIST has relocated the use guidelines from Section 3 to a web-based format, structured as an interactive FAQ page to improve usability and facilitate timely updates in response to user needs. NIST is currently encouraging the public to review the draft update and submit feedback, with comments accepted until June 13, 2025. After considering the feedback, NIST anticipates releasing a final version of the updated framework later this year.
Meanwhile, the Bluetooth Special Interest Group (SIG) has also made strides in enhancing privacy with the release of Bluetooth 6.1 as part of its new bi-annual update schedule. This update introduces a feature called Randomized Resolvable Private Address (RPA), designed to enhance both privacy and power efficiency. The RPA feature makes tracking Bluetooth devices significantly more challenging by randomizing the timing of device address changes rather than using predictable intervals.
This update follows Bluetooth 6.0, which arrived in September 2024 and introduced Channel Sounding for centimeter-level location accuracy, lower latency, and faster scanning capabilities. The upcoming iPhone 17 series, expected to be released in September, will likely support Bluetooth 6.0 and possibly incorporate the new features from version 6.1. With Bluetooth now on a twice-yearly release schedule, the next update, Bluetooth 6.2, is expected to arrive by the end of 2025 with additional refinements and features.
As organizations and tech companies navigate the complexities of privacy management, the importance of effective governance structures cannot be overstated. In Canada, privacy laws continue to evolve, necessitating that organizations build governance frameworks that meet new legal and operational expectations. Experts like Tricia Ralph and Kaitlyn Hebert explore strategies for establishing privacy programs that support compliance and accountability, emphasizing the need for organizations to stay ahead of legal developments.
This comprehensive approach to privacy, from consumer technology to organizational governance, reflects a growing recognition of the importance of protecting personal data in an increasingly digital world.
