The Shift from Hand to Face Is Changing How We Interact With Technology
I. The Migration of the Interface
Meta’s Ray-Ban glasses let a wearer ask a spoken question about whatever is in front of them, translate a foreign phrase in real time, or get spoken directions, and hear the answer through open-ear speakers without a phone ever leaving a pocket. That small absence of a familiar gesture is the story.
For roughly fifteen years, personal computing has increasingly settled into a familiar mechanical ritual: an individual experiences an impulse, reaches into a pocket or bag, extracts a rectangular glass slab, unlocks the screen, casts their eyes downward, executes a series of taps and swipes, and finally returns their gaze to the physical world. This sequence has become so routine that it is frequently mistaken for the permanent state of mobile technology.
Yet this interaction model imposes real friction. The smartphone often demands manual and visual attention that pulls a person’s focus away from their immediate surroundings, even for a few seconds.
The important shift is not that glasses are replacing phones. It is that computing is moving from being a destination to being an ambient layer around the wearer. A smartphone is inherently passive, resting in fabric until explicitly summoned. Glasses, positioned at the primary sensory nexus of the human body, perceive the external world alongside the wearer, so a question can be asked and answered without a separate device ever being retrieved. Every other formulation in this article, that the screen becomes optional, that the phone becomes a hub rather than an interface, that the face is becoming the new site of interaction, is a consequence of that one shift rather than a separate idea competing with it.
This evolution is advancing through measurable, discrete commercial stages rather than an immediate leap into science fiction. The first stage, already operating at meaningful scale, dispenses with visual displays entirely, combining microphones, open-ear speakers, cameras, and multimodal AI within conventional optical frames. The second stage introduces heads-up micro-displays. Full spatial augmented reality remains an emerging horizon beyond that. Tracing this progression shows that computing is not simply undergoing another cycle of miniaturization; it is relocating where human-machine interaction takes place.
II. Why Earlier Smart Glasses Failed
Cranial computing is not a new idea. Over the past decade and a half, the consumer electronics industry made several high-profile attempts to put computers on people’s faces. Google Glass debuted in 2013 as a conspicuous, prism-display computing experiment worn openly on the face, and drew immediate public controversy over its always-present camera, with businesses and individuals objecting to being recorded without warning. Google ultimately shifted Glass away from the mainstream consumer market toward enterprise applications. Snap’s Spectacles, beginning in 2016, were a camera-first device built primarily around capturing short social video, giving them a narrower everyday utility than later AI-enabled glasses. Meta’s first-generation Ray-Ban Stories, released in 2021 with EssilorLuxottica, adopted a familiar eyewear form factor but offered only limited camera and audio functions with no real contextual AI, and did not reach the commercial scale of the AI-enabled generation that followed it.
There was no single cause of their limited consumer adoption. Several factors compounded: conspicuous, unfamiliar designs that read as computing hardware rather than eyewear; genuinely limited day-to-day utility; privacy and social friction created by a face-mounted camera, as Glass’s own backlash demonstrated; battery and component limitations that forced trade-offs in comfort or capability; and, underlying all of it, an unclear case for why an ordinary consumer needed the product at all.
One useful way to read that history, offered here as this article’s interpretation rather than an established historical fact, is that these devices lacked contextual intelligence. Mounting a low-resolution camera, a small prism display, and a micro-battery onto a frame produced hardware without a correspondingly compelling reason to wear it. The interface was clunky, relying on side-frame taps or rigid voice commands, and the device had no meaningful understanding of what the wearer was actually looking at or doing.
What changed with the current generation was primarily the arrival of multimodal AI models capable of seeing, hearing, and reasoning about a scene together. Pairing that software with efficient sensors gave this generation of eyewear a much more coherent operational rationale distinct from an awkward mini smartphone screen: a physical-world sensory terminal for an assistant that could actually understand the world in front of it. Ray-Ban Meta, launched by Meta in 2023, became a prominent commercial example of this shift as Meta subsequently added multimodal AI that could interpret what the wearer was seeing, turning the camera and audio hardware into a functioning assistant rather than a recording accessory.
III. The Screenless Breakthrough
A useful way to understand the transition is as a continuum, and one of the first commercially significant stages of it did not arrive by placing high-resolution displays in front of human eyes. It arrived by removing the display altogether.
Market research from Omdia illustrates the pace of this shift: global shipments of AI glasses reached 4.2 million units in the first half of 2026, a 127 percent increase year-over-year, with Meta accounting for roughly 3.4 million of those units and an estimated 81.3 percent share of the global AI-glasses market over that period.
IDC’s Q2 2026 shipment data divides the broader category into three groups. The category labels and shares below follow IDC; the input and output columns are this article’s synthesis:
| Hardware category | Share | Primary inputs | Primary outputs |
|---|---|---|---|
| Display-less audio/AI glasses | 70.3% | Beamforming microphones, integrated camera | Open-ear directional speakers |
| Display glasses | 14.3% | Microphones, camera, some with neural wristband | Micro-display (monocular or binocular), audio |
| XR headsets | 15.4% | Eye tracking, spatial cameras, hand gestures | Waveguide or opaque displays, spatial audio |
Display-less glasses currently dominate shipments, although shipment share alone cannot establish why consumers choose them. One plausible product logic, offered as editorial synthesis rather than a documented engineering fact, is a chain: eliminating the display can reduce weight and complexity; a lighter design can help preserve a familiar eyewear silhouette; a familiar silhouette may reduce the visual signal that someone is wearing technology; and an AI system can still perform many useful tasks through audio and a camera without a display.
Pricing history is worth stating plainly rather than blending across generations: Ray-Ban Meta launched at $299 in 2023, while the current Ray-Ban Meta Gen 2 starts at $379. Meta has also introduced a separate Meta Glasses line starting at $299. By combining open-ear audio, directional microphones, an integrated camera, and multimodal AI, display-less glasses demonstrated a commercially viable form of this computing layer. Ray-Ban’s own product documentation describes wearers capturing photos and video, making calls, streaming music, translating spoken language in real time, and asking questions about objects in their line of sight. The first stage of this platform shift is defined less by what the wearer can see through the glasses than by how much a screen turned out not to be necessary.
IV. The Sensory Hierarchy: Eye, Ear, and Voice
Moving computing from the hand to the face reorganizes the interface itself. Under this article’s own framework, that reorganization can be described as a shift from comparatively allocentric to egocentric sensing: allocentric meaning the device observes the world from an external vantage point the user must consciously aim, and egocentric meaning the device shares the wearer’s own point of view and moves with it.
A smartphone camera fits the allocentric side of that framework. It captures only what the user deliberately points it at, typically from waist or chest height. A camera mounted beside the eye is egocentric: it shares the wearer’s visual field and follows natural head movement without being aimed at all. Ray-Ban’s own documentation describes the multimodal system interpreting objects, text, landmarks, and mechanical assemblies that fall within that field of view, feeding aspects of the surrounding environment into the assistant’s contextual understanding.
Acoustic design plays an equally important role, even though the outward-facing camera draws most public attention. Ray-Ban Meta glasses use open-ear speakers and multiple microphones to direct sound toward the wearer while reducing outward leakage and separating the wearer’s voice from background noise. That combination enables a conversational interface in which natural speech, rather than typing, becomes the primary way of issuing a request. Google has described spoken dialogue as the central interaction model for Android XR in its own developer materials.
Each modality in this hierarchy contributes something distinct:
| Modality | What it enables |
|---|---|
| Camera / vision | Understanding of the physical scene |
| Microphones | Hearing and situational context |
| Speakers | Low-visibility audio responses |
| Voice | Explicit, spoken intent |
| Gesture / EMG | Silent, low-motion intent |
| Display | Optional visual output |
| Spatial tracking | Spatial registration for anchored digital content |
The important transition running through all of it is bidirectional context: the glasses perceive the environment and can return information without requiring the wearer to reach for a separate device. That is a meaningfully different relationship between person and computer than tapping symbols on a handheld pane of glass.
V. The Return of the Display and the Neural Band
Display-less designs opened the initial consumer market, but the category is now expanding to include visual information as well. In September 2025, Meta introduced the Ray-Ban Meta Display, priced at $799 at launch including the Neural Band, marking a second commercial stage for face-based computing, according to Meta’s newsroom.
Audio alone works well for a specific set of tasks: translation, reminders, direct answers, and turn-by-turn prompts. A display becomes valuable once information is spatial, visual, persistent, or simply hard to convey through speech, such as live captions during a conversation, translated text, a map, or an incoming message a wearer wants to read rather than hear. Framed that way, the display’s return is an earned response to a specific limitation of audio rather than an inevitable next step for its own sake. The Ray-Ban Meta Display integrates a compact visual display directly into the lens, letting the system show navigation prompts, message previews, live captions, and AI answers in the wearer’s peripheral field.
A display, however, reintroduces a familiar interaction problem: voice commands are not always practical in crowded, quiet, or noisy environments, whether that means a commuter train, an office meeting, or a factory floor. Meta’s answer, available as part of the Ray-Ban Meta Display system, is the Neural Band, a wrist-worn accessory using surface electromyography, or sEMG. It does not read brainwaves. It attempts to turn subtle wrist and finger muscle signals into a control channel that does not require the wearer to speak, touch a phone, or make a large visible gesture. According to Meta’s developer documentation, the sEMG sensors can detect small finger movements with the hand resting on a table or inside a pocket, letting a wearer dismiss a notification, scroll text, or confirm an action with a barely visible motion.
What is currently demonstrated is a narrow set of these gestures, mostly navigation and confirmation actions, built into the Display system. The broader idea of a general-purpose muscle-signal input layer for computing is a future possibility the current hardware gestures toward, not a capability that exists today. Meta has begun making the platform available to developers through its Developer Preview, creating a path for broader experimentation with the interaction model.
VI. Competing Ecosystems
The growth of smart eyewear has set off a platform contest between vertically integrated hardware makers and more open software platforms. Google’s response is Android XR, a platform developed with partners including Samsung and Qualcomm to provide a shared software foundation across optical and spatial devices, according to Google’s own materials. Google separates the category into audio glasses, built around open-ear audio and conversational AI, and display glasses, which add a heads-up visual layer for navigation, translation, and richer applications.
A significant part of Google and Samsung’s approach is partnering with established eyewear brands rather than launching frames as standalone gadgets: Samsung’s own announcements describe collaborations with Gentle Monster and Warby Parker. For a mass-market product, appearance and familiarity become unusually important because the device is worn continuously on the face, which is a large part of why earlier attempts struggled. Meta’s own relationship with EssilorLuxottica, the parent of Ray-Ban and Oakley, is the clearest working example: it gives Meta access to Ray-Ban’s brand recognition, EssilorLuxottica’s optical manufacturing capacity, and an existing retail and prescription-eyewear network that a technology company could not easily build on its own.
It is useful to separate the competition into three layers rather than treating Google, Meta, Apple, and Snap simply as rival hardware companies:
- The operating system and AI layer, where Android XR competes with Meta’s own AI and Horizon platforms.
- The hardware and optics layer, covering cameras, waveguides, batteries, and chips.
- The eyewear, distribution, and fashion layer, covering frame design, retail presence, and brand trust.
On developer tooling specifically, Google brings an established Android application ecosystem that third-party developers can extend into eyewear, while Meta is building a dedicated wearables developer stack around its Wearables Device Access Toolkit and, for display glasses, Web Apps, rather than relying solely on its existing immersive-device development ecosystem.
VII. The Full Augmented Reality Horizon
Audio glasses and display glasses represent the near-term commercial center of the category, but the technological horizon is full augmented reality. Snap’s standalone Specs, priced at $2,195 for developers and enterprise customers, with shipments planned in the United States, the United Kingdom, and France, illustrate this tier, according to Snap’s own announcements.
What separates full AR from display AI glasses is not simply “more advanced hardware.” It requires several things at once: spatial registration of the physical room, persistent digital objects that stay anchored in space as the wearer moves, depth and spatial understanding of the environment, low-latency rendering, optics capable of showing bright graphics without excessive power draw, and interaction models built for three-dimensional space rather than a flat screen. Specs include dual see-through waveguide displays, SLAM cameras for real-time mapping, and hand-tracking sensors that together meet that bar in a way audio and display glasses do not attempt to.
At $2,195, Specs are positioned far above conventional smart glasses, and Snap has focused on enterprise use, with Reuters reporting commercial partnerships with Salesforce, Nvidia, and AWS. Snap’s own materials describe enterprise applications spanning retail floors, factories, and field operations, illustrating a broader pattern worth stating as this article’s own synthesis: full AR may find its earliest economic footing in settings where hands-free visual information has clear, immediate operational value, well before consumer pricing makes it a mass-market product. These technologies are easier to understand as stages in a continuum than as competing categories, which raises a broader question about how the whole category fits together.
VIII. The Continuum of Wearable Computing
A useful way to understand this transition is as a continuum of hardware and interface types rather than a single replacement event for the smartphone. Framed this way, and offered here as an interpretive model rather than an official industry taxonomy, each stage represents a distinct computing paradigm:
- Smartphone: destination computing, where the user leaves their physical environment to visit an app inside a handheld screen.
- Screenless AI glasses: ambient contextual computing, where the assistant works as a background sensory layer delivered through audio.
- Display glasses: heads-up computing, adding a persistent but limited visual layer to that same ambient assistant.
- Spatial AR: immersive spatial computing, where digital objects are anchored directly into the physical environment.
The point of this continuum is the interface transition it describes, not simply a list of devices getting more capable. Where a factor such as “social friction” appears in a comparison, it should be read as an editorial assessment rather than a published industry metric: a smartphone carries low social friction because raising it to use it is a recognized, expected gesture, while a face-mounted camera carries meaningfully higher friction because there is no equivalent visible cue that recording might be happening. Wearable eyewear functions as the practical bridge between today’s ambient audio assistants and a more fully spatial computing future.
IX. Silicon Bottlenecks and Why the Smartphone Survives
A common claim in coverage of this category is that smartphones are becoming obsolete. The available evidence does not support that. A more accurate description is that the smartphone is being redefined rather than replaced: it is ceasing to be the exclusive interface and settling into a role as a computational and connectivity hub.
The mechanism behind that persistence is physical. Glasses sit on delicate facial structure, so weight remains a major constraint: the device has to carry electronics, batteries, and increasingly optical components while staying light enough for comfortable all-day wear. That weight limit in turn constrains how much battery the frames can carry, which limits how much continuous processing they can sustain on their own. Limited battery and processing capacity encourage designs that rely on a paired phone or cloud infrastructure for computationally heavy AI, while current lightweight glasses often depend on another device for broader connectivity. The practical result is greater reliance on a paired phone or cloud infrastructure for more demanding functions, while the glasses handle the contextual sensing and interaction closest to the wearer. Ray-Ban Meta’s own product documentation, for instance, requires the glasses to be paired with a phone and companion app for setup and several connected functions, illustrating that the glasses remain part of a larger computing system rather than a completely self-contained computer.
Chip designers have built specialized silicon to work within that constraint. Qualcomm’s Snapdragon AR1 Gen 1 platform is built around integrated AI processing, a capable image signal processor for camera work, and power and thermal optimization aimed at features like on-device voice detection, according to Qualcomm’s own product materials, while heavier workloads can still be handled by a paired phone or cloud service. Optical engineering compounds the challenge: reporting from Reuters on the long-running collaboration between Applied Materials and EssilorLuxottica describes waveguide design that preserves outdoor visibility without excessive battery drain as one of the harder open problems in the category. The resulting arrangement, as reflected in Google and Samsung’s own platform materials, is complementary rather than a clean replacement: the phone often remains the higher-capacity computational and connectivity hub, while the glasses serve as the contextual front end, the eyes, ears, and voice, of the same system.
X. Specialized Utility: Translation, Accessibility, and New Markets
For smart eyewear, some of the clearest documented use cases are those that solve a specific problem more conveniently than a phone, and two examples demonstrate that clearly: real-time translation and accessibility.
Translation removes a specific piece of smartphone friction. Instead of taking out a phone, opening an app, and holding it toward a speaker, Meta’s own documentation describes glasses capturing spoken dialogue through their microphones, translating it, and either playing the result into the wearer’s ear or showing captions on display-equipped models, all without a device ever coming out of a pocket.
Accessibility is a second documented case. Meta has added functionality built on its partnership with Be My Eyes, along with its own visual-assistance features, giving blind and low-vision wearers hands-free help identifying objects, reading text, and navigating their surroundings, according to Meta’s newsroom. That illustrates a broader possibility worth stating plainly as a possibility rather than a settled outcome: a phone can also provide visual assistance, but normally requires the user to hold and aim it, while a camera-equipped AI assistant worn on the face can turn a wearer’s surroundings into an accessible information layer without that step.
Meta has also rolled out region-specific updates in India, including Hindi-language conversational support, according to its own announcements. That is the language currently documented; claims about Tamil or Bengali support are not established by current evidence, so as support expands to more Indian languages the potential use cases could extend further, without assuming they already have.
The reason this market may matter in particular is worth stating directly: in settings where hands are often occupied, a smartphone screen is inconvenient to use, and voice interaction is genuinely practical, translation and voice-first assistance could be especially useful. That is a plausible use case, not a demonstrated pattern of adoption, and the distinction is worth holding onto.
XI. The Privacy Asymmetry
Alongside its utility, facial computing raises a real problem for bystander privacy, and the clearest way to state it is a simple asymmetry: the wearer chooses to wear the device. The people around the wearer do not choose anything.
That asymmetry breaks into two separate questions. User privacy covers what happens to the wearer’s own recordings, location, voice, queries, AI conversation history, and account data, which raises many of the same data and consent questions that apply to other connected devices. Bystander privacy covers something different: what happens to people who never opted into the interaction at all, people captured incidentally in a frame, and any facial or identifying information swept up along with them.
A person raising a camera or phone to eye level normally creates a visible recording cue, giving bystanders a chance to notice, object, or step out of frame. Glasses make that cue less reliable: a wearer simply looking around is physically indistinguishable from a wearer using the camera, and on newer Meta models some camera-based AI functions do not trigger the capture LED at all.
Manufacturers have tried to compensate with hardware signals. Ray-Ban Meta glasses include a white capture LED that signals when photos or videos are being captured for the gallery. Meta says the light cannot be disabled and that newer models automatically disable the camera when they detect that the LED has been blocked or tampered with. But the privacy mechanism is more limited than a simple ‘camera active’ indicator: Ray-Ban’s current documentation says that on newer models, AI features that use the camera, such as identifying a landmark or plant, do not turn on the capture LED because that imagery is not intended for sharing.
Whether that indicator actually does its job in practice is a separate question. Its effectiveness depends on several things holding up at once: how visible it is in bright outdoor light, where a small LED can be hard to see; how resistant it is to tampering, since attempts to cover or modify the light have prompted Meta to strengthen its tamper-detection measures; how consistently Meta’s own software enforcement catches a covered or removed light; and how much bystanders themselves understand what the light means in the first place. That is the real mechanism behind the privacy concern, not a vague sense of unease: a notice system whose practical value depends on visibility, tamper resistance, and public awareness working together.
The potential risk was illustrated by Harvard’s I-XRAY project in late 2024, which showed how off-the-shelf smart glasses could be combined with facial-search services and public data brokers to identify strangers and surface personal information about them in something close to real time. Concrete institutional responses have followed rather than one sweeping continental rule. His Majesty’s Courts and Tribunals Service has banned Meta’s smart glasses from courts across England and Wales, confiscating them at the door because existing rules already prohibit recording inside a courtroom; Oslo has banned the glasses in its schools; and Reuters reported in September 2026 that Australia is considering barring camera-equipped smart glasses from government workplaces on similar grounds. These individual restrictions illustrate a broader institutional concern about face-mounted cameras rather than a single unified legal response, and the England and Wales example makes the underlying concern tangible: a device that looks like ordinary eyewear can be prohibited precisely because its camera is difficult for anyone else in the room to notice.
One proposed response to camera-related backlash has been camera-free glasses that retain microphones, open-ear audio, and conversational AI while removing the camera, according to September 2026 reporting from Reuters. That tradeoff is significant: removing the camera also removes the AI’s ability to see what the wearer sees, reducing the system’s contextual capabilities in exchange for removing the camera-related privacy concerns that accompany visual sensing, and it does not resolve the separate question of ambient audio recording, since microphones alone still raise their own consent questions. A related and unresolved debate concerns cloud versus on-device processing: cloud-based models can be larger and more capable, but require sending raw sensory data off the device, while local processing increases the amount of sensitive processing kept on the device itself and reduces, without eliminating, dependence on cloud transmission. OpenAI CEO Sam Altman has argued, as Reuters reported in September 2026, that building consumer trust in wearable AI will increasingly depend on doing more of that sensitive processing locally.
The privacy problem and the potential commercial opportunity described next arise from the same underlying capability: the glasses can sense aspects of context that make them useful, and that same contextual sensing could become commercially valuable.
XII. The Commercialization of Visual Attention
Beyond the immediate recording question, facial computing raises a longer-term question about the commercial value of attention itself. A search engine records what someone types. An e-commerce site tracks what someone clicks.
An egocentric system built into glasses works at a different level: rather than recording an explicit query, an egocentric system could potentially capture information from the wearer’s visual and auditory environment. With outward cameras and, potentially, gaze tracking, such a system could in principle register what a wearer looks at in a store, how long attention lingers on a billboard, or what places and objects someone regularly encounters. The chain from that observation to a commercial outcome runs through several distinct, non-obvious steps: the camera sees something, the AI interprets what it saw, the platform could infer a context or interest from that interpretation, and only after all of that could a platform potentially deliver a commercial response. That last step remains hypothetical. No major glasses maker has confirmed a gaze-based advertising product, and this section describes a possible future business model, ambient contextual advertising, rather than something that currently exists.
The incentive for building it is nonetheless real: Reuters recently described the smartphone market as roughly a $600 billion market, making even a partial shift of economic activity toward a new interface strategically significant. Control of the facial interface could give a platform unusually intimate access to physical-world context, which is what makes ownership of that interface strategically valuable to the companies pursuing it. Establishing clear boundaries around visual telemetry and biometric data retention therefore raises significant consumer-protection questions as the category develops.
XIII. The Screen Becomes Optional
The shift from hand to face is not, at bottom, a story about chips, silicon nodes, or waveguides. It is a story about where a person’s attention goes.
The smartphone freed computing from the desk, but it also created a familiar physical pattern: people repeatedly looking down at a screen while moving through the physical world, dividing attention between what is in front of them and what is in their hand. Facial computing offers something different: the possibility of engaging with digital information without looking away from the people and places actually in front of the wearer. Meta has framed its Ray-Ban Display around a similar idea, with Mark Zuckerberg describing glasses as a form factor that lets people stay present while still drawing on AI, and Snap has used comparable language in describing its own Specs launch.
That promise carries its own tension. As this kind of computing becomes less visible, it also becomes harder to notice from the outside, which is exactly the problem explored in Section XI: the same design choices that make glasses comfortable and unobtrusive to wear are what make it difficult for a bystander to tell when a camera is active.
That is the full shift this article has traced: the smartphone made computing a destination; multimodal AI made glasses useful enough to justify wearing them; screenless audio glasses became the first widely commercialized form of the new interface; and the display returned once visual information earned its added complexity. Augmented reality extends that trajectory into spatial computing, while physical limits keep the phone as a hub, and the same sensing that enables translation and accessibility creates new privacy and commercial questions. None of that means glasses replace phones. It means the screen is losing its monopoly over how people reach a computer.
References
- Google Blog: Intelligent Eyewear with Gemini Is Coming This Fall (Android XR Overview)
- IDC: The Rise of Smart Glasses, From Novelty to Necessity
- Omdia: Global AI Glasses Shipments More Than Doubled in 1H26, Hitting a Record 4.2 Million
- IDC: Smart Glasses Growth Cools to 35% as Meta Doubles Down on Distribution (Q2 2026 Market Data)
- Ray-Ban: Ray-Ban Meta Smart Glasses FAQs & Features
- Meta Newsroom: Ray-Ban Meta (Gen 2) Available Now, Starting at $379
- Meta Newsroom: Meta Ray-Ban Display, AI Glasses With an EMG Wristband
- Meta Developers: Build for Display Glasses (Meta Horizon OS)
- Samsung Newsroom: Samsung and Google Give First Look at New Intelligent Eyewear
- Snap Newsroom: Introducing Specs Augmented Reality Glasses
- Snap Inc. Investor Relations: Specs Make Computing More Human with New Experiences, Partnerships, and Specs Intelligence
- Meta Newsroom: New Meta AI Features Make Ray-Ban Glasses in India More Useful
- Be My Eyes: Meta AI Glasses FAQ and Accessibility Integration
- Meta Newsroom: Advancing Accessibility at Meta
- Reuters: From Smart Glasses to AI Pins, Privacy Fears Challenge Tech’s Next Big Bet (September 2026)
- Reuters: Australia Considers Banning the Use of Smart Glasses in Government Buildings (September 17, 2026)
- The Guardian / HMCTS: Courts in England and Wales Ban Meta’s Smart Glasses (August 2026)
- Reuters: coverage of camera-free smart glasses and on-device AI processing, including comments from OpenAI CEO Sam Altman (September 2026)
- Harvard I-XRAY Project: original 2024 demonstration combining smart glasses with facial-search services
- Reuters / Investing.com: EssilorLuxottica, Applied Materials Strike Deal to Develop Smart Glasses AR Technology
- Android: Learn About Extended Reality with Android XR
- Reuters: Snap Targets Enterprises with Salesforce, Nvidia AI Tools for Specs AR Glasses
- Meta Newsroom: Partnering With EssilorLuxottica to Launch Meta Glasses
- Qualcomm Technologies, Inc.: Snapdragon AR1 Gen 1 Platform Architecture
