Meta has now lost more than $60 billion on its Reality Labs division since 2020. That's not a typo. Sixty billion dollars — roughly the GDP of Luxembourg — poured into headsets that most people own but don't use, and glasses that may or may not change everything depending on who you ask.
Enjoying this? Never miss a story.
This week, Meta's Q1 2026 earnings dropped another Reality Labs loss of $4.2 billion in a single quarter, bringing the year off to its now-familiar, eye-watering start. The stock barely flinched. Which tells you something important about how Wall Street has decided to interpret Mark Zuckerberg's most expensive obsession.
Introduction
The phrase "Meta is burning money on AR/VR" has appeared in headlines so many times it's starting to feel like a seasonal event — like pumpkin spice lattes, but for investor anxiety. And yet, here we are again, with Reality Labs posting losses that would make most Fortune 500 CFOs need a paper bag.
What's different about this moment is that the narrative is actually starting to split. On one side: the critics who have been right about the metaverse being a ghost town (remember Horizon Worlds? Tumbleweeds.) On the other: a growing number of analysts who think Meta's Ray-Ban smart glasses — which sold over 10 million units in 2025 — represent something genuinely new happening beneath the headline losses.
So is this a company throwing good money after bad, or the most patient long-game bet in Silicon Valley history? That question is worth taking seriously, because the answer affects not just Meta's stock price but the entire trajectory of how we interact with computing over the next decade. Here's what's actually happening.
The Numbers That Keep Getting Worse (or Do They?)
Let's start with the raw financials, because the scale genuinely matters. Reality Labs — the division that houses Meta Quest headsets, Ray-Ban Meta smart glasses, and all the metaverse infrastructure — has lost $58.1 billion cumulatively between 2020 and the end of 2025, according to Meta's own earnings filings.
For context: that's more than Apple spent developing the original iPhone over its entire pre-launch R&D period, multiplied by a factor that should make anyone uncomfortable.
Meta's core business — Facebook, Instagram, WhatsApp — generated $164.5 billion in revenue in 2025, so the company can technically afford this. The advertising engine is printing money. But "can afford it" and "should be doing it" are different questions, and Meta's shareholders have been asking the second one loudly since 2022.
Why the Stock Stopped Punishing Them
Here's the interesting part. Meta's stock hit an all-time high of $741 in February 2025, even as Reality Labs losses continued. That's because Wall Street quietly made a deal with Zuckerberg: keep the ad business growing, keep the cost cuts you made in the 2022-2023 "Year of Efficiency," and we'll tolerate the AR/VR bonfire.
It's a fragile truce. If ad revenue slows — say, because of a U.S. recession or a significant TikTok resurgence — Reality Labs becomes the first thing institutional investors will demand gets cut. That's the structural risk no earnings call addresses directly.
What Meta Actually Announced vs. What Actually Happened
In October 2021, Zuckerberg renamed Facebook to Meta and announced the company was going all-in on the metaverse. The pitch: a persistent, immersive virtual world where people would work, socialize, and spend money. They announced a future where the physical and digital would merge seamlessly.
Here's the reality: Horizon Worlds, Meta's flagship metaverse platform, peaked at around 300,000 monthly active users in early 2023 — a number that a moderately successful indie game would be embarrassed by. The avatars didn't have legs until late 2022 (the jokes wrote themselves). The whole thing felt like a tech demo that someone forgot to finish.
The Quest headsets have fared better. The Meta Quest 3, released in October 2023 at $499, is genuinely the best standalone mixed-reality headset on the market — better optics, better performance, and a more compelling use case than Apple's Vision Pro at $3,499. (The Vision Pro is an extraordinary piece of hardware that almost nobody can justify buying. That's a separate article.)
But Quest unit sales have plateaued. Meta doesn't release exact numbers, but third-party estimates from IDC put standalone VR headset shipments at around 7.7 million units globally in 2025 — a market that's growing, but slowly. "Mass adoption" this is not.
The Ray-Ban Story Is the One Actually Worth Watching
Buried under the metaverse discourse is the product that might actually matter: the Ray-Ban Meta smart glasses, now in their third generation.
The original Ray-Ban Stories launched in 2021 to mild curiosity and privacy concerns. (A tiny LED indicating recording was on. People covered it with tape. Classic.) The second generation, released in late 2023, added a live AI assistant — ask it what's in front of you, translate a sign, identify a plant — and something clicked.
Sales accelerated sharply. By the end of 2025, Meta confirmed crossing 10 million units sold for the Ray-Ban line. That's not iPhone numbers, but it's the first AR-adjacent wearable from any major tech company to achieve genuine consumer traction. Google Glass sold fewer than 100,000 units. Snap's Spectacles remain a developer curiosity.
Why the Glasses Actually Work (When the Headsets Didn't)
The answer is almost embarrassingly simple: you don't have to put something on your face that makes you look like a deep-sea diver. Ray-Ban Meta glasses look like sunglasses. People wear sunglasses. Problem solved.
The form factor insight — that AR's first mass-market moment would come through fashion, not technology — is something Meta got right and almost everyone else got wrong. It's also why Apple's Vision Pro, despite its technical brilliance, is struggling to find a mainstream identity. Nobody wants to strap a ski goggle to their face on the subway. (They just don't.)
The fourth-generation Ray-Ban Meta glasses, expected in late 2026, are rumored to include a small display — an actual heads-up display in a frame that looks normal. If Meta pulls that off at a reasonable price point, the calculus on all those losses starts to look different.
Is This a Problem? Depends on Who You Ask.
Critics will point out — correctly — that Meta has been "almost there" on AR/VR for five years now, and the goalposts keep moving. The metaverse was supposed to be the future of work by 2023. It wasn't. Headsets were supposed to hit mainstream adoption by 2024. They didn't. Smart glasses with displays were rumored for 2025. They're now a 2026 story.
But here's why that framing misses the point: the timeline criticism assumes there's a specific deadline by which this needs to work. There isn't. Meta is not a startup running out of runway. It's a company generating $40 billion in annual free cash flow that has decided AR/VR is the next major computing platform and is willing to subsidize that bet for as long as the core business holds.
The more honest question is whether the bet is structurally sound — and that depends on whether you believe spatial computing is a real paradigm shift or an expensive hobby for tech executives who got bored of flat screens.
What the Analysts Are Actually Saying
Mark Mahaney at Evercore ISI, one of the sharper tech analysts on the Street, wrote in a March 2026 note that Reality Labs losses are "a known and priced-in risk" but flagged that the division needs to show "a credible path to revenue contribution by 2027 or investor patience will erode." That's the professional way of saying: the clock is ticking, but it hasn't run out yet.
On the other side, Gene Munster at Deepwater Asset Management has been bullish on Meta's AR trajectory, arguing that the Ray-Ban glasses represent "the beginning of a 10-year platform transition" comparable to the shift from desktop to mobile. He's been saying this for a while, so take it with appropriate salt. But he's not wrong that mobile computing looked like a toy market in 2005.
The Competitive Landscape Nobody Talks About Enough
Meta's AR/VR spending doesn't happen in a vacuum. Apple, Google, Samsung, and a constellation of Chinese manufacturers are all playing in this space — and the competitive dynamics matter for whether Meta's $60 billion bet pays off or gets outflanked.
Apple's Vision Pro launched in February 2024 at $3,499 and has sold an estimated 500,000 to 700,000 units through early 2026, according to analyst estimates — far below Apple's typical launch trajectories. A lower-cost "Vision" model, rumored at under $1,500, is expected sometime in 2026. If Apple cracks the price barrier, Meta's Quest line faces a serious challenger from a company with better brand trust in consumer hardware.
Samsung's Galaxy XR headset, developed in partnership with Google and Qualcomm, launched in mid-2025. It's a credible device. It hasn't moved the needle on adoption either. Which suggests the problem isn't Meta-specific — the entire category is still looking for its iPhone moment.
Meanwhile, companies like Snap are taking a different approach — leaning hard into developer-facing AR tools and the Spectacles platform, betting that AR's first killer apps will come from developers rather than hardware manufacturers. They may be right. They may also be running out of money before they find out. (Snap's stock is down 68% from its 2021 peak. Just noting that.)
What This Means for the Average Person Right Now
If you're not an investor in Meta and not a developer building XR apps, you might reasonably ask: why does any of this matter to me? Fair question.
Here's the practical takeaway: the AR/VR battle happening right now at the corporate level will determine what your next computing interface looks like in 5-10 years. Every dollar Meta, Apple, and Google are spending on this today is R&D for the device you'll buy in 2030.
If you're mildly curious about the space without wanting to spend $500 on a Quest headset, the Ray-Ban Meta glasses (starting at $299) are the most accessible on-ramp right now. They're useful as actual sunglasses, the camera is solid, and the AI assistant is genuinely handy in ways that are hard to predict until you use them. That's not an ad — it's just the honest state of the consumer AR market in 2026.
Also worth reading: The App Putting AI Characters in Your Group Chat — because the line between AR overlays and AI-enhanced communication is blurring faster than most people realize. And Snapchat's New Ad Strategy Wants to Have a Conversation With You is relevant context for how social platforms are rethinking the interface layer entirely.
The Bottom Line
Meta is still burning money on AR/VR. That part of the headline is accurate. But "burning money" implies waste, and the picture is more complicated than that. The metaverse gambit — the Horizon Worlds, the legless avatars, the virtual office meetings nobody wanted — that part has largely failed, and Meta has quietly deprioritized it without ever formally admitting defeat. (The company calls this "refocusing." What it actually did was pivot away from its biggest public embarrassment.)
What's replaced it is more grounded: a headset business that's the clear market leader in standalone VR, and a smart glasses line that has genuine consumer momentum for the first time in the category's history. Neither is a sure thing. But neither is the reckless money pit that 2022 headlines suggested.
My read: Meta will not abandon this bet. Zuckerberg has staked too much personal credibility on spatial computing, and the Ray-Ban sales give him real ammunition to keep the board patient. The next 18 months — specifically whether the fourth-gen glasses ship with a working display and whether a credible "killer app" emerges for Quest — will determine whether this goes down as the most expensive pivot in tech history or one of the most vindicated. I'm not saying it'll work. I'm saying it's not over, and the people declaring it dead are about three product cycles too early.