NeuroAI NEUROAINEUROAI.SITE
ESC

The Headset Didn't Sell, So It Became a Screen

At a launch event in late September, Meta showed a pair of glasses. There is no strap. It weighs about 100 grams. The computing, battery, and storage have been moved into a tethered puck that clips

2026-11-05 · 5973 words · NeuroAI
The Headset Didn't Sell, So It Became a Screen

At a launch event in late September, Meta showed a pair of glasses. There is no strap. It weighs about 100 grams. The computing, battery, and storage have been moved into a tethered puck that clips to a pocket or bag.

The device is called Meta VR Glasses. It goes on sale next spring at $1,299.99. It connects to virtual reality on one side and to Meta's own AI agent on the other, like a seam between two products.

In the same quarter's earnings, the division that makes this hardware brought in just $443 million, down 6 percent from a year earlier, with an operating loss of $4.13 billion. The two figures sit awkwardly together.

The division's annual revenue is about $2 billion, roughly 1 percent of the company's total. It looks less like a printing press than a laboratory that has been set apart from the rest of the business.

A Small Decimal in the Earnings

$443 million is a rounding error next to Meta's $64.6 billion in quarterly revenue. The loss that came with it was $4.13 billion, more than nine times the revenue.

Meta's explanation is simple: Quest headsets sold less and dragged revenue down; AI glasses sold well and offset part of it. Add and subtract, and the result is still a subtraction, because the hole left by headsets is bigger than the gain from glasses.

The company also gave guidance: Reality Labs is expected to post an operating loss of $18 billion to $19 billion this year, narrowing to around $15 billion next year. That is the company's own range, not an outside estimate.

At $19 billion, the division burns more than $50 million a day. A day, not a year. Selling glasses will not close that gap anytime soon, and cost-cutting will not either. This is an old bill in the hardware business.

Headsets Stall, Glasses Sell Out

The two product lines are moving in opposite directions. Headsets are retreating; glasses are advancing. IDC data shows global smart-glasses shipments of 9.6 million units last year, with Meta accounting for about 76.1 percent, a share rarely seen in hardware.

The same firm expects the category to reach 13.4 million units this year, up nearly 40 percent, fast for consumer electronics. The glasses are still glasses; they simply no longer require putting a person inside a room.

The partner's earlier figure was that Ray-Ban Meta sold more than 7 million units last year, more than three times the prior year. One pair of glasses has carried the division's face, and a shape that does not look strange outdoors is worth more than a pile of specs.

Seven million sounds like a lot, but spread across the world it is still a small category. It is far from one pair per person, and also far from unsellable. In between sits a market that is still taking shape.

How Ninety Billion Piled Up

Add up the operating losses in the annual reports, and from 2020 to today the division has lost more than $90 billion. This is not one year's story but six years', and there has not been a single positive year.

In January, the division cut about 10 percent of its staff, more than 1,000 positions. Three studios making virtual-reality games were shut. In late February, enterprise Quest headsets and managed services stopped being sold.

Put together, the moves point to one thing: the center of gravity is shifting from putting people into a virtual world to putting AI on people. The company's phrase in its earnings was reallocating investment to other priority areas.

The money saved from layoffs was said to go into wearables. The same hands, moved from headsets to glasses. The sentence carries no emotion, but it is clearer than any slogan: the direction has changed.

The New Glasses Are Not the Next Quest

On specs alone, the new glasses resemble a shrunken headset: 5K resolution, micro-OLED panels, a 70-degree field of view, much narrower than the company's own Quest 3. What was given up in field of view was traded for weight, a clear trade-off.

But it is defined as something else. Meta put its AI agent directly into the glasses' operating system, controlled by voice, eye movement, and gestures, with no controller. What it sells is a large screen you carry with you; the glasses are only the shell.

Some media described the step as "VR demoted to a display for AI." That is a media reading, not Meta's own wording. Meta's phrasing is that the device carries "personal superintelligence," and the media prefers the first phrase because it is shorter.

The two descriptions point to roughly the same picture: you used to put on a headset and enter a world; now you put on a pair of glasses and let AI hand you a screen. The difference is only who is doing the talking.

The Money Turns Toward the Data Center

Meta's capital expenditure guidance this year is $135 billion to $145 billion, with the lower bound raised by $5 billion. The money goes mainly to data centers, chips, and networks.

On one side is a hardware division with a few hundred million in revenue and billions in losses; on the other is a hundred-billion-dollar scale of compute investment. Put the two ledgers together and the priority needs no explanation; the numbers speak for themselves.

Smart glasses are growing quickly, but their unit price is low and the supply chain is long. They cannot fill the hole left by headsets in a short time. The phrase "offset part of it" in the earnings means exactly this, and this is the hardest part of the hardware business.

Capital expenditure and divisional losses are two separate ledgers and cannot be subtracted directly. But place them on the same page and the direction is clear: money is moving toward the data center, hardware is moving toward lightness, and the two lines are drifting apart.

Look at the glasses and the Quest together, and they answer different questions. The Quest asks whether you want to enter another world. The glasses ask whether you want to wear one more pair of glasses.

Ten Years of a Screen

More than a decade ago, virtual reality was pitched as the next computing platform. The imagined future was that people would put it on and work, socialize, and watch movies inside it, and screens would disappear. The demo videos of that era were rough.

More than a decade later, the screen has not disappeared. It has been moved in front of the eyes. What changed is the position, not the role. The headset retreated from an entrance to a display, and the screen moved from a window to the lead.

This is not only Meta's story. The whole industry is turning heavy head-mounted devices into glasses you can wear outside. IDC says headset shipments fell sharply last year while glasses rose, two curves crossing, and the supply chain is changing with them.

A screen can be a good business. Monitors have been sold for decades and have never been considered obsolete. Their story just does not sound as grand as a platform, and does not promise as much.

The 100-gram glasses do not go on sale until next spring. Whether they sell will depend on whether someone who watches a whole movie in them wants to put them on a second time. The second time is the real demand; wearing them the next year is how they stay.

Disclaimer: This article is based on publicly available information and is for reference only. It does not constitute investment advice.

Related coverage

Cite this article

NeuroAI (脑机网), "The Headset Didn't Sell, So It Became a Screen," 5 November 2026, https://neuroai.site/a/na-meta-vr-26-1105

Machine-readable: BibTeX · RIS · JSON · Markdown

More in “Companies & Stack” → · Back to home · Markdown version