AI Glasses Grew 263%. Meta Owns Almost All of It.

Global shipments of AI glasses rose 263% year over year in the first half of 2026, according to Counterpoint Research. That is the headline. The structure underneath it is more interesting than the growth rate.

Display-less glasses — frames with cameras, microphones, speakers and an assistant, but no screen in the lens — made up 96% of everything shipped. Within that segment, Meta held 94%. Xiaomi and Alibaba, in second and third, held 1% each.

The other 4% of the market is glasses with a display in the lens, which Counterpoint groups as AR and AI glasses. That segment grew faster, 449% year over year, and it has a different leader: Rokid at 41%, Meta at 37%, Even Realities at 11%, Alibaba at 7% and INMO at 2%.

So the category now has two maps. On one, Meta is effectively the market. On the other, it is a close second to a Chinese company most American buyers have never heard of.

The display-less market is a Meta product line

A 94% share is not a competitive market with a leader. It is a product line with a few rounding errors next to it.

The reasons are not mysterious. Meta chose the form factor early, when the rest of the industry was still chasing headsets and waveguides, and it chose it with EssilorLuxottica, the partner that owns the frames people already wear. The Ray-Ban and Oakley lines put the hardware into an existing retail channel and an existing fashion decision; buyers are not choosing a computer for their face, they are choosing sunglasses that happen to take photos and answer questions. The assistant does not need to be the best on the market. It needs to be good enough and already on your head.

The regional split shows how concentrated the demand still is. North America took 50% of display-less shipments and Western Europe 28%. China, the world's largest consumer-electronics market, accounted for 3%. This is, for now, a Western category built around one brand.

That concentration comes with a cost Meta is carrying alone. The glasses have drawn a steady run of criticism over their cameras and what it means to record strangers without an obvious signal. When one company holds nearly the whole segment, that criticism does not land on "smart glasses." It lands on Meta, and the category's reputation becomes a function of a single company's privacy record.

The display market is where the fight is

The AR and AI slice is a small fraction of units — roughly one in twenty-five glasses shipped — but it is the part of the category where the product is still being decided.

Here the geography flips. China took 45% of shipments, North America 41% and the rest of the world 14%. Rokid's lead reflects a domestic market that has been willing to buy display glasses at volume while Western buyers have mostly waited. Even Realities, at 11%, is the pure-play bet that a minimal monochrome display for notes, translation and navigation is worth more than a camera.

Meta's 37% here rests largely on the Ray-Ban Display, its first consumer glasses with a screen in the lens. Second place in a segment it entered recently is a respectable position. It is also a signal that the playbook that works for display-less frames — brand, retail, good-enough software — does not transfer automatically once the lens has to render something.

Last week Snap began selling its standalone Specs at $2,195, a full AR computer in the frame with no tethered puck. That is a different product from anything in the display-less market, at several times the price of a pair of Ray-Ban Metas. It will not move Counterpoint's unit shares much. It does show where the high end of the display segment is heading.

Why it matters

Three things follow from these numbers.

The form factor question is settled for now. For years the industry argued about whether face computing would arrive as a headset or as glasses. At 96% of AI-glasses volume, the answer for the mass market is glasses without a display — an assistant with eyes and ears, not a screen. Displays are growing faster from a smaller base, but the thing people actually buy today is closer to earbuds with a camera than to a phone on your face.

Meta has a real platform, which means it has a platform problem. Holding 94% of a hardware segment gives Meta the installed base to make its assistant the default interface for ambient, camera-first AI. It also makes Meta the only company whose behavior determines whether regulators and venues treat the category as acceptable. A single high-profile privacy failure would hit the entire market at once.

The competitive window is short. Counterpoint expects the next phase to be "agentic" glasses that act more proactively on what they see and hear, and new form factors aimed at privacy, including camera-free audio glasses. Google and Samsung are preparing Android XR glasses that will launch display-less first, tied to Gemini and the phone. Apple is widely reported to be targeting 2027 for its own display-less glasses. Each of those companies brings a phone ecosystem Meta does not have, and the ecosystem — messages, maps, payments, the assistant that already knows your calendar — is what turns a camera on your face into something you rely on.

Meta's advantage is that it got there first and the others are arriving to a market it already owns. Its disadvantage is that the others own the phone in your pocket, and glasses without a display depend on that phone more than Meta would like to admit.

For now, the numbers say something simple. The category is real, it is growing fast, and for everyone except Rokid, the competition is less about beating Meta than about deciding what Meta's glasses will have to become.