Remote assistance, AI-guided work and hands-free capture are three of the leading enterprise applications. But buyers still need clear answers on hardware, trusted knowledge, privacy and fleet control.
Enterprise AI glasses have an evidence problem. Buyers can already identify plausible applications, but most can’t point to scaled deployments backed by a year of return-on-investment data.
ArborXR says companies are still mainly running proofs of concept. Interest is clustering around three workflows:
- Connecting a worker to a remote expert
- Providing AI-guided instructions
- Capturing images or video without reaching for another device
AI glasses are one part of the wider smart glasses market. The term is appropriate here because the applications rely on AI assistance, cameras, audio and access to information, including on devices without a display.
ArborXR’s earlier roundtable with The XR Beat explored lighter hardware and the controls needed around privacy, identity and data. A follow-up discussion with ArborXR Co-Founder and Head of Product Jordan Williams and Devin Marble, Manager Strategic Partnerships, examined what companies can deploy now and what still needs proving.
These Three Workflows Are Leading Enterprise Tests
Remote Assistance
A technician, inspector or field engineer can share a first-person camera feed with a specialist who sees the equipment and speaks to the wearer. Both hands remain available for the job.
One organisation described to ArborXR sends workers across the United States to assess equipment for resale. They then inspect damage and condition in environments where safety-rated glasses may be required. Remote assistance could allow a specialist to review an unusual asset without travelling to every inspection.
Devin Marble, Manager Strategic Partnerships explained:
“You’ve got to have your hands free. Time is very important, but what does it take to save that person time? It’s knowledge. It’s some sort of subject-matter expertise, whether that’s AI or a human.”
Any saving must still be measured. A pilot should record how often the expert resolves the issue, how much travel is avoided, task completion time and sessions disrupted by connectivity or hardware problems.
AI-Guided Work
AI glasses can retrieve procedures, explain fault codes, identify parts or adapt instructions to the worker’s experience.
An insurance adjuster examining rare property damage is one example being explored. Hands-free capture can feed evidence into an existing system, while an approved knowledge source tells the adjuster what to inspect.
The quality of that knowledge is critical. The system must understand the company’s equipment, terminology and procedures.
Co-Founder and Head of Product Jordan Williams went on to say:
“How do we ensure that the knowledge being surfaced, or the two-way interaction that the frontline worker is having with the AI, is accurate and reliable and they can really trust it? It needs to know all the weird random acronyms of the protocols and whatnot.”
Companies need to identify the knowledge source, its owner and its update process. They must also decide when the system can answer directly, when it should cite a source and when the worker needs a human expert.
Hands-Free Capture
A mechanic can document a component, an inspector can record an asset’s condition and a technician can capture compliance evidence without putting down tools.
Cameras and microphones introduce governance requirements. Buyers need rules covering consent, storage, retention and access. Camera-equipped and camera-free devices also present different workplace risks, as explored in The XR Beat’s analysis of smart-glasses privacy and product categories.
Remote Assistance: A Faster Route to a Live Pilot?
Remote assistance can reach a live test sooner than a bespoke guided-work or training application because it does not require a library of 3D content. The initial workflow needs a camera, two-way audio, an available expert and a defined support process.
“Even if they’ve already selected their content, deployment and headset, there’ll probably be a minimum of 90 days before anyone is actually using it. You hand them glasses and they’re managed and there’s remote assist working: you’re using them on day two.”
Devin Marble, Manager Strategic Partnerships
The comparison is an illustration, not a universal deployment promise. Security review, network access, worker consent, device pairing and support procedures can extend the timeline. Remote assistance can still test a live job without first commissioning a complete instructional application.
ArborXR currently supports point-of-view video and two-way audio on selected Meta AI glasses. The paired Android phone is enrolled in ArborXR, while the glasses appear in the portal as a peripheral.
The distinction is important. IT may be managing two connected endpoints, with separate decisions covering phone ownership, user accounts, Bluetooth pairing, charging and updates. Seeing glasses in a management portal doesn’t mean administrators have the same controls available on a managed Android headset or industrial wearable.
Google also lists ArborXR as an Enterprise Mobility Management partner for Android XR, among the validated management solutions supporting Android Enterprise features for XR. The controls available will still depend on the device and how it connects to the platform.
ArborXR’s early projects are primarily using company-owned, shared devices. Shared use introduces further questions around sign-in, prescription requirements, cleaning, resetting settings between users and handling captured data at the end of a shift.
A New Market With Promising Initial Results
Because the market is still young, ArborXR doesn’t yet have a mature, named smart-glasses case study with fleet size, cost and measured return ready for publication.
Marble said:
“The returns that people are seeing today are the beginning of time savings. I’m very excited to learn what the actual return on their investment is going to be. We’ll probably need at least a year to be able to truly measure it.”
The lack of mature evidence should shape the pilot. Buyers can establish a baseline before devices arrive and measure:
- Task and resolution time
- First-time completion rate
- Expert travel and call-outs avoided
- Errors, rework and incomplete records
- Device uptime and support tickets
- Worker acceptance after the novelty period
Hardware price is only part of the calculation. Paired phones, connectivity, licences, support, cleaning, replacement stock, knowledge preparation and systems integration also belong in the cost base.
The XR Beat’s smart-glasses pilot guide provides five questions buyers can use before selecting hardware.
Hear the Deployment Detail on 30 September
ArborXR’s webinar, A New Wave of Wearables Is Coming. Is Your Company Ready?, takes place on 30 September 2026 at 1pm ET / 10am PT / 6pm BST / 7pm CEST.
Devin Marble, Jordan Williams and Jason J. Lee will examine the devices reaching the market, current workplace applications and the privacy, support and fleet-control decisions that emerge as pilots grow.
