The workplace has spent the past decade making communication more portable. Meetings moved from boardrooms to laptops, calls moved from desk phones to mobile devices, and artificial intelligence is now beginning to sit inside the tools employees use every day.
AI glasses are the next, slightly more complicated, step.
They could give workers spoken updates before a meeting, connect field teams with remote experts and make information available without another screen getting in the way. For employees moving between a desk, a customer site and a physical task, that is a genuinely compelling idea.
But the same device that can offer useful assistance can also introduce a new route for sensitive information to leave the workplace. A camera can capture a confidential design. A microphone can pick up a private conversation. An AI assistant can process information that was never intended to reach a third-party platform.
That is why restrictions on AI glasses are starting to emerge in the types of spaces where communications and data need the tightest controls. New York’s Unified Court System banned smart glasses from court facilities in July, including eyewear or headwear that contains cameras, microphones, computers or technology capable of recording video or audio. Courts in England and Wales have also prohibited Meta smart glasses, while Australia is considering restrictions on camera-equipped glasses in government workplaces.
These measures might sound far removed from everyday business life. They are not. They reflect a question that enterprises will increasingly have to answer in offices, hospitals, manufacturing sites, laboratories and customer-facing environments: when a wearable can hear, see and connect to AI, what rules should apply?
Meta’s new Ray-Ban Meta Audio glasses offer one possible answer. By removing the camera, the company is testing whether an audio-first AI wearable can be more acceptable in the places where camera-equipped glasses are likely to face the greatest resistance. It is a useful case study, but it also shows that removing one sensor does not remove the wider governance challenge.
TL;DR
Camera-equipped AI glasses could create privacy, security and workplace-recording risks in sensitive business environments.
A blanket ban may be necessary in some spaces, but it could also block useful hands-free communication, accessibility and frontline support use cases.
Meta’s Ray-Ban Meta Audio glasses remove the camera, but microphones and AI access mean enterprises still need clear rules around data, consent and identity.
Why Are AI Glasses So Difficult to Govern at Work?
Let’s be honest, the issue is not just the technology itself. We have had cameras, microphones and cloud software in our pockets for years.
The difference is visibility.
If someone holds up their phone in a meeting, everyone knows what is happening. If somebody wears ordinary-looking glasses with a camera, microphone and AI assistant built in, the line becomes much less clear. Colleagues may not know whether they are being recorded, whether a conversation is being transcribed, or whether an AI service is processing what is being said.
That creates a headache for businesses that have spent years building controls around their collaboration platforms. UC teams can set policies for meeting recordings, manage user identities, apply retention rules and monitor access to enterprise applications. Personal AI glasses can sit outside those controls, especially if they connect through a consumer account or companion app.
Workplace guidance from Fisher Phillips highlights the potential for AI glasses to capture proprietary processes, trade secrets, internal conversations and customer or patient information. It also notes that information could be transmitted to third-party AI platforms outside an employer’s direct control.
That is the real UC concern. A wearable is not just another gadget if it can capture communications, interpret them and act on them. It is another endpoint. And endpoints need governance.
What Is an AI Glasses Endpoint?
An AI glasses endpoint is a wearable device that can capture or deliver communications through microphones, speakers, cameras, voice assistants, cloud connectivity or AI. Enterprises should assess it by what it can hear, see, process and access, not by the fact that it looks like a normal pair of glasses.
Where Could Camera-Equipped AI Glasses Be Restricted?
Not every workplace will need to ban AI glasses, some spaces carry obvious risk. A product-development lab can expose confidential designs, while healthcare environments may involve patient information. Manufacturing sites can reveal proprietary processes and safety controls, and contact centres handle customer data at scale. Even a routine client meeting may cover commercially sensitive plans that should not be recorded, analysed or stored outside approved systems.
In those environments, the camera is the most obvious concern. It can capture information quickly, discreetly and at scale. That makes a camera-equipped pair of glasses fundamentally different from a device that only delivers audio.
There is still more to consider, though. Microphones can be just as sensitive when confidential meetings, private customer conversations or regulated communications are involved. And AI features add another layer of complexity. Where does the information go? Is it retained? Is it used to train a model? Can an administrator see it? Can a worker connect the glasses to company data?
The policy cannot just say “no smart glasses.” That is far too vague. Businesses need to decide which features are restricted, which spaces are restricted and why.
What Could Businesses Miss by Banning AI Glasses?
This is where the conversation becomes more interesting. A full ban may feel like the simplest response, but it could also mean missing out on genuinely useful workplace technology.
For frontline workers, hands-free access to information could be valuable. A field engineer could hear instructions while working on equipment. A warehouse employee could receive safety prompts without reaching for a handheld device. A healthcare worker could potentially access information without breaking focus during a task.
For office employees, the opportunity is slightly different. An AI assistant could provide a spoken briefing before a meeting, remind someone about an overdue action or help them contact a colleague while they are moving between spaces. It is not about replacing a laptop. It is about making the screen optional for smaller, everyday tasks.
There is also an accessibility argument. Emerging AI wearables may support real-time transcription, object recognition, audio prompts and navigation assistance. That means employers should not dismiss every request to use the technology as a personal preference. Some cases will need a proper, individual assessment.
That does not mean organisations should open the door to every device. It means a ban should be based on the risk, rather than a fear of anything new.
