On March 18, 2026, Meta sent an email to Horizon Worlds users telling them that their virtual reality social platform was being shut down. By June 15, the company said, the app would no longer function on Quest headsets. It would exist only as a mobile app. The announcement landed with relatively little public mourning — a comment section here, a Reddit thread there — which itself said something about how far the platform had fallen from its original ambitions.
Then, within 48 hours, Meta reversed the decision.
Meta’s Chief Technology Officer Andrew Bosworth appeared in an Instagram Stories Q&A and told followers that the company had decided to keep Horizon Worlds running in VR. “We have decided, just today in fact, that we will keep Horizon Worlds working in VR,” Bosworth said, after a user wrote in to say they were “heartbroken” about the shutdown.
What happened in that 48-hour window matters less than what the whole sequence reveals about where Meta’s massive metaverse experiment actually stands right now. Behind both the announcement and the reversal sit five years of accumulating losses, a user base that never grew to meaningful scale, a global VR hardware market in consecutive decline, and a company that has spent $83.6 billion — with a cumulative operating deficit that by some estimates now approaches $90 billion — building something that most consumers decided they did not want or need.
This post covers what actually happened with the Horizon Worlds VR shutdown, why the reversal does not change the underlying picture, what the financial data shows, and what the whole episode means for businesses, marketers, and anyone tracking how major technology platforms live and die.
What Meta Actually Announced on March 18, 2026
Meta posted a message to its community forums on March 18, 2026, stating that Horizon Worlds would be removed from the Quest store at the end of March and shut down entirely in VR by June 15. After that date, the platform would operate only as a standalone mobile application. Meta’s stated reason for the change was a desire to give each version of the platform more focused development.
“We are separating the two platforms so each can grow with greater focus, and the Horizon Worlds platform will become a mobile-only experience,” Meta said in the community announcement.
The shift came weeks after Meta had already cut over 1,500 employees from Reality Labs, the division responsible for VR, augmented reality, and metaverse development. Among those cuts were teams working on Horizon Worlds itself, including Ouro Interactive, an in-house studio that had been created in 2023 specifically to build first-party content for the platform.
The news was not entirely surprising. Meta had been signaling its reduced commitment to VR-as-metaverse for over a year. By late 2024, the word “metaverse” had quietly disappeared from Meta’s earnings calls entirely. In February 2026, Reality Labs VP of Content Samantha Ryan had already told developers the company would be “doubling down on the VR developer ecosystem while shifting the focus of Worlds to be almost exclusively mobile.” The March announcement formalized that direction for users.
What did surprise observers was the speed and apparent finality of the move — and then its swift reversal.
The 48-Hour Reversal: What Andrew Bosworth Said
Two days after the original shutdown announcement, Bosworth took to Instagram Stories to walk back the most significant part of it. He said Meta had decided to keep Horizon Worlds functional in VR “for the foreseeable future.” His explanation cited user feedback directly: “the fans who reached out,” he said, were the reason the company changed course.
A Meta spokesperson confirmed Bosworth’s comments to TechCrunch.
The framing of the reversal deserves scrutiny. Bosworth’s statement preserved the core shift in strategy: Meta is not resuming active development of new VR Horizon Worlds content. The mobile version remains the company’s primary focus. All that changed was that existing VR worlds would not be removed. Users who want to continue visiting the spaces they built or enjoyed in VR can still do so — but those spaces will receive no new content, no updates, no developer investment. The VR version of Horizon Worlds is, in practical terms, being preserved rather than supported.
Bosworth had made the company’s thinking plain in an earlier podcast conversation with journalist Alex Heath. He explained that the development team had been forced to build everything twice — once for mobile, once for VR — and that the drain on engineering resources was untenable.
“There’s a much bigger audience in mobile, and it’s having a really positive pickup on mobile,” Bosworth said. “They’re having to build everything twice — they’re building it once for the mobile phone, and they’re building it again for VR. There’s a pretty easy way to increase their velocity, which is just like, let them build for mobile.”
The reversal also put a spotlight on the speed with which Meta had made and then walked back a major platform decision. The company announced a full VR shutdown, received public criticism, and reversed within 48 hours — a sequence that raises questions about how well these decisions were reviewed before announcement, and what it signals about the internal state of Meta’s VR operations.
A Platform That Was Never Able to Find Its Audience
To understand the context around the Horizon Worlds situation, you need to understand how it got here. Horizon Worlds officially launched to the public in late 2021, the same year Meta changed its name from Facebook. The renaming was a statement of direction. Mark Zuckerberg described the metaverse as “the next frontier,” predicting it would reach a billion people and generate “hundreds of billions of dollars of digital commerce” within a decade.
The platform had serious problems almost from day one. User avatars launched without legs, an omission that attracted widespread mockery and became part of a broader cultural backlash against Meta’s metaverse vision. Zuckerberg’s own pixelated avatar became a meme. That tone — one of skepticism shading into ridicule — never fully lifted from how the public perceived Horizon Worlds.
The user numbers told a corresponding story. By October 2022, just months after the platform’s wider public launch, Horizon Worlds had fewer than 200,000 monthly active users. This fell short of an internal target of 280,000 — a target that had itself been revised down from an earlier goal of 500,000. At no point did the platform exceed a few hundred thousand monthly active users.
That ceiling matters in context. Roblox, a platform Horizon Worlds was often compared to, was reporting over 65 million daily active users by 2025. VRChat, the scrappier, user-run social VR platform that pre-dated Horizon Worlds by years, maintained a more engaged and organic user community. Meta had far more marketing spend, far more distribution reach through its existing platforms, and far more financial resources than any competitor. None of those advantages translated into an audience.
The platform attracted partnerships with major brands and artists. Concerts from Imagine Dragons and Coldplay were staged in virtual spaces. Company after company set up branded virtual experiences inside Horizon Worlds, investing budget into activations for an audience that was simply not there at meaningful scale. The problem, as Forrester VP Mike Proulx put it plainly to Wired: “Advertisers follow their target audiences. And those audiences were never inside Horizon Worlds.”
The Financial Reality: $83.6 Billion in Reality Labs Losses
The shutdown of Horizon Worlds VR cannot be separated from the financial story of Reality Labs, the Meta division responsible for it. The numbers here are significant enough to require close attention.
Reality Labs has posted an operating loss every single year since Meta began reporting it as a separate segment in 2020. Those losses have grown every year. In 2025, the division reported an operating loss of $19.1 billion — up from $17.7 billion in 2024. That annual loss came against full-year revenue of just $2.2 billion for the division, a revenue-to-loss ratio that is stark by any measure.
In the fourth quarter of 2025 alone, Reality Labs posted an operating loss of $6.02 billion on revenue of $955 million. That single quarter’s loss represented roughly 3% of Meta’s entire company revenue for the period, drawn from a division that generated approximately 1% of the company’s total income.
The cumulative total of these losses reached $83.6 billion through 2025, according to Meta’s earnings reports. Some analysts tracking the full picture, accounting for additional spending categories, place the figure closer to $90 billion. To put that in the terms TechCrunch used: you would need to spend $1 million per day for 200 years to spend $73 billion. The actual figure is higher.
These losses are not a secret. Zuckerberg has been transparent with investors about them, framing them as long-term investment spending. But the gap between what has been spent and what has been returned — in users, in revenue, in any conventional measure of platform success — has widened, not narrowed, over the five-year period.
Meta’s overall business remains healthy. The company reported $200.97 billion in total revenue in 2025, up 22% year-over-year. That financial strength is the only reason the Reality Labs experiment was able to continue at all. But investor patience with the losses has its limits, and the layoffs, studio closures, and platform pivots of early 2026 all point to a recalibration.
The Layoffs: What the Reality Labs Cuts Actually Involved
The Reality Labs workforce reductions in January 2026 were the most significant in the division’s history, but they did not arrive without context. Meta cut approximately 21,000 employees across the whole company during 2022 and 2023, a period when the stock fell sharply and investor pressure mounted. Reality Labs saw additional cuts in mid-2024. The January 2026 reduction of approximately 1,500 Reality Labs employees — about 10% of the division — came on top of those prior rounds.
What was distinctive about the January 2026 cuts was their specificity. Teams working directly on Horizon Worlds were hit hard. Several in-house game studios were shuttered entirely, including Ouro Interactive. The Workrooms app, Meta’s VR meeting platform aimed at enterprise users, was retired. Third-party development of Horizon OS headsets was paused. These were not general cost reduction measures — they targeted the specific infrastructure Meta had built to develop and populate the metaverse.
Reports in March 2026 also indicated that Meta was considering a further, more significant round of layoffs that could affect up to 20% of the company overall. No announcement had been made as of publication, but the possibility underscored the degree to which Meta is actively restructuring its cost base as it redirects investment toward artificial intelligence.
CNBC has reported that the January 2026 Reality Labs cuts sparked concern in the broader VR industry about a potential “VR winter” — a period of contracted investment and reduced activity similar to the AI winters of past decades. Whether that concern proves justified depends significantly on how other hardware players, including Valve, Samsung, Apple, and Pico, execute in the space during the remainder of 2026.
Why VR Failed to Scale: The Hardware Problem
The Horizon Worlds story is partly about a social platform that failed to attract users. But it is also about a hardware constraint that no amount of software investment could overcome.
Virtual reality headsets require users to put on a wearable device that most people, in practice, find uncomfortable for extended use, socially awkward to wear, and expensive to purchase. The Meta Quest 3 launched at $499. Apple’s Vision Pro launched at $3,499 and Apple subsequently had to scale back production due to disappointing demand. Neither device became the kind of commodity hardware that social platforms need to reach broad consumer audiences.
The numbers at the market level are unambiguous. Global VR headset shipments fell 12% in 2024, marking the third consecutive year of decline. In the first half of 2025, shipments fell another 14% year-over-year. Meta’s Quest headset sales specifically dropped 16% from 2024 to 2025, according to IDC data. By IDC’s forecast, full-year 2025 headset shipments were set to decline 42.8% compared to prior projections.
These figures matter for understanding the Horizon Worlds situation because even a perfect social VR platform — engaging, polished, safe, and free — would have struggled to grow under these hardware adoption constraints. You cannot build a mass social platform that requires a device most households do not own. Forrester’s Proulx made this point directly: “Meta was trying to solve for a consumer problem that doesn’t exist. You can’t build a mass social platform reliant on hardware most people neither own nor want to wear for more than short bursts.”
This was a structural problem, not a product problem, and it was identifiable from the beginning. The user growth ceiling Horizon Worlds hit in 2022 reflected it. The platform’s dependence on Quest headsets meant its potential audience was always the subset of people who had purchased that device, not the billions of social media users Meta needed to replicate the scale of Facebook or Instagram.
The Mobile Pivot: What the Numbers Show
The shift from VR to mobile for Horizon Worlds was not an emergency pivot made in response to the March 2026 announcement. Meta launched a mobile version of Horizon Worlds in September 2023, explicitly to reach users who did not own VR headsets and to build a footprint that more closely resembled Roblox’s browser-and-mobile model.
The mobile version has shown more traction than the VR version — but it is important to distinguish between download figures and meaningful engagement. According to mobile intelligence firm Appfigures, the Horizon Worlds mobile app had accumulated 45 million total worldwide downloads across iOS and Google Play as of March 2026. Downloads in 2026 to date were approximately 1.5 million, representing a 53% year-over-year increase compared to the same period in the prior year.
Those download numbers sound substantial. The revenue number tells a different story. Total consumer spending on the Horizon Worlds mobile app has reached $1.1 million in aggregate. That figure represents the sum of all in-app purchases made by all users across all markets since the mobile version launched. Against the backdrop of $83.6 billion in Reality Labs losses, $1.1 million in consumer revenue is essentially a rounding error.
The gap between download volume and revenue generation suggests the mobile app is attracting curious users who are not converting into paying customers at any meaningful rate. Whether that changes as Horizon Worlds receives dedicated mobile development resources — now that the team is no longer splitting effort between VR and mobile — remains to be seen. Bosworth has described the mobile product as having “really positive pickup,” but the financial data does not yet validate that characterization at scale.
Meta’s Actual Success: Ray-Ban Smart Glasses
In the same period that Horizon Worlds was struggling to hold a few hundred thousand users, a different Meta hardware product was quietly building genuine momentum.
Meta’s Ray-Ban smart glasses — designed in partnership with EssilorLuxottica and running AI-powered features through the companion app — sold over 7 million units in 2025 alone. That represented a tripling of sales from the prior year, confirming the product’s trajectory rather than hinting at it. EssilorLuxottica’s shares hit an all-time high as investors responded to the sales data.
The Ray-Ban smart glasses product works because it fits into an existing consumer behavior — wearing glasses — rather than requiring consumers to adopt an entirely new interaction paradigm. It sits on the nose, looks like an ordinary pair of sunglasses or frames, and integrates AI assistants that can identify objects, play music, answer questions, and take photos. The device does not ask users to wear a headset, sit in front of a computer, or enter a virtual world.
Zuckerberg has noted that Ray-Ban smart glasses sales tripled year-over-year, and has used the product as evidence that Meta’s hardware strategy is not failing uniformly — it is succeeding in form factors that make behavioral and commercial sense. The company is also pursuing Oakley-branded smart glasses through the same EssilorLuxottica partnership, targeting a different demographic.
The Ray-Ban success stands in significant contrast to the Quest headset trajectory. It suggests Meta’s hardware instincts are sound when applied to devices people actually want to use — and that the company’s mistake with the metaverse was not a failure of hardware engineering but a failure of understanding what users would willingly integrate into their lives.
The AI Pivot: How ChatGPT Changed Meta’s Priorities
Understanding what happened to the metaverse at Meta requires placing it in the context of what happened to artificial intelligence — not just at Meta, but across the entire technology industry.
ChatGPT launched publicly in November 2022. Its reception was unlike almost anything the technology industry had seen. Within months, every major technology company was rearranging priorities, reallocating budgets, and publicly reorienting their strategies around AI. Meta was no exception.
In the months following ChatGPT’s launch, Meta accelerated its public pivot toward AI. The company released the Llama family of open-source large language models. It integrated AI assistants into WhatsApp, Instagram, and Facebook. It announced Meta AI as a standalone product. It committed to $135 billion in capital expenditure for 2026, a significant portion of which is directed at AI infrastructure including data centers and chips.
By the Q3 2024 earnings call, Meta’s executives had stopped using the word “metaverse” entirely. The quarterly reports, once anchored by Zuckerberg’s elaborations on the metaverse roadmap, shifted to discussions of AI model performance, AI assistant engagement, and AI-powered advertising optimization. Reality Labs continued to be reported as a division, and its losses continued to be disclosed, but the strategic narrative around it had changed.
Bosworth’s current description of the “metaverse” — as a concept encompassing AR, digital overlays on physical spaces, and even using smartphones in shared physical environments — represents a significant definitional shift from the 2021 vision of virtual reality social worlds that would replace screens and physical space. The word is being retained while its meaning is being quietly amended.
Meta’s $135 billion AI investment plan for 2026 dwarfs Reality Labs’ entire historical loss total. It signals where the company’s ambition now sits.
What This Means for Brands and Marketers
The Horizon Worlds story carries direct lessons for any business that invested budget in metaverse marketing, metaverse brand experiences, or any emerging platform whose audience remained theoretical.
The first lesson is about hardware dependency. Any platform whose reach is defined by ownership of a specific device carries a ceiling that no amount of marketing investment can raise. When brands built Horizon Worlds experiences during 2022 and 2023, the addressable audience was limited to Quest headset owners — a group that never numbered in the tens of millions, and that was actively shrinking as a percentage of the consumer population as VR headset shipments declined for three consecutive years. The reach numbers for branded metaverse campaigns were always optimistic.
The second lesson is about verifying active user data before committing to a platform. Download counts, registered accounts, and even reported monthly active user figures can obscure the reality of genuine engagement. Horizon Worlds’ reported user figures were regularly revised downward. The mobile app’s 45 million downloads coexist with $1.1 million in total consumer spending — a combination that should prompt questions about the quality of that engagement before any brand invests meaningfully in it as a platform.
The third lesson is about platform continuity and commitment. Meta announced a full VR shutdown and reversed it within 48 hours. Even if the platform technically survives in VR, no new development is planned, the team has been significantly cut, and the company’s strategic focus has moved on. Brands that built in Horizon Worlds for the long term made their bets on a platform that the platform owner had already decided to deprioritize. The pace at which major technology platforms shift their strategic commitments has accelerated in the AI era.
The fourth lesson, perhaps the most important one, is about audience-first platform selection. Forrester’s Proulx noted that advertisers follow their target audiences, and that those audiences were simply never inside Horizon Worlds. The corollary is straightforward: before committing marketing budget to any new platform or format, the first question is not “is this platform interesting?” but “are my actual customers using this?”
The Broader VR Market in 2026: Context Matters
The Meta Horizon Worlds story does not mean virtual reality as a technology is finished. But the market data from 2024 and 2025 paints a picture of an industry that has not yet found the product-market fit that would drive mass adoption.
Global VR headset shipments fell 12% in 2024, the third consecutive year of decline. In H1 2025, shipments fell another 14% year-over-year. IDC forecast a 42.8% decline in full-year 2025 headset shipments compared to earlier projections. Apple’s Vision Pro, positioned as a premium spatial computing device rather than a gaming headset, had production scaled back due to disappointing demand.
New hardware is entering the market in 2026. Valve has a Steam Frame headset in development. Samsung launched its Galaxy XR device in December 2025 running the Android XR platform. ByteDance-owned Pico plans to release its Project Swan headset later in 2026. Apple’s Vision Pro received an M5 chipset upgrade. Meta has stated it maintains “a robust road map of future VR headsets tailored to different audience segments.”
Whether any of this hardware moves the needle on consumer adoption depends on factors that have remained stubbornly resistant to marketing and investment: price, comfort, content ecosystems, and compelling use cases that justify the device’s presence in a consumer’s life. The Horizon Worlds failure does not prove VR cannot succeed. It does demonstrate that a social VR platform with few hundred thousand peak users is not the use case that gets it there.
What Meta Says It Is Doing Next
In Meta’s public statements, the company continues to position itself as the largest investor in VR globally, even as it restructures Reality Labs and moves Horizon Worlds to mobile-only development.
Zuckerberg has outlined a two-track strategy. One track is AI glasses and wearables — building on the Ray-Ban success with more capable devices that integrate AI assistance into everyday life without requiring immersive VR. The company is developing glasses that can, as Zuckerberg has described, “see, hear, and generate custom interfaces” as an ambient layer over the physical world.
The second track is making the VR hardware business profitable over a longer time horizon, with no specific timeline attached. Zuckerberg told investors that Reality Labs losses in 2026 are expected to remain similar to the 2025 figure of $19.1 billion, and that this year represents the peak before reductions begin. Whether losses actually decrease in 2027 and beyond will depend on whether the wearables strategy generates meaningful revenue at scale.
The Horizon Worlds mobile app is being positioned as the consumer-facing social platform product within this strategy — a Roblox-style mobile experience that can grow independently of VR hardware adoption. The 53% year-over-year download growth in early 2026 is being cited as evidence of that potential. The $1.1 million in lifetime consumer spending on the app is not.
Meta’s description of the metaverse has also quietly shifted to something far broader than it was in 2021. Bosworth now describes the concept as encompassing augmented reality, digital overlays on physical spaces, and even the use of smartphones in shared physical environments. Under that definition, almost any mobile or digital experience qualifies as part of the metaverse. The definitional expansion appears designed to allow the company to claim metaverse progress without building the immersive VR social worlds that were originally promised.
The Record in Plain Terms
Here is what the available evidence shows, stripped of narrative framing on either side:
Meta spent $83.6 billion in cumulative Reality Labs operating losses between 2020 and 2025. Horizon Worlds, the flagship consumer product of that investment, peaked at fewer than 200,000 monthly active users in 2022 and never recovered meaningful growth. The VR headset market declined for three consecutive years. Meta laid off more than 1,500 Reality Labs employees in January 2026, shuttered dedicated game studios, and moved Horizon Worlds development to mobile. The company announced a full VR shutdown of Horizon Worlds, reversed the decision within 48 hours in response to user feedback, and confirmed no new VR development would follow. The mobile app has 45 million downloads but $1.1 million in lifetime consumer spending. The one hardware product showing genuine traction — Ray-Ban smart glasses — runs on AI, not VR. The company’s declared investment priorities for 2026 center on artificial intelligence at a scale that dwarfs its entire metaverse spend.
That record does not require dramatic interpretation. It is what it is.
Frequently Asked Questions About Meta Horizon Worlds, the VR Reversal, and Reality Labs Losses
What exactly happened with the Horizon Worlds VR shutdown?
On March 18, 2026, Meta announced through its community forums that Horizon Worlds would be removed from the Quest VR headset store at the end of March and would be fully shut down in VR on June 15, 2026. After that date, the platform was to exist only as a mobile app. Within two days, Meta CTO Andrew Bosworth reversed the announcement during an Instagram Stories Q&A session, stating that Meta had decided to keep Horizon Worlds running in VR “for the foreseeable future.” The reversal was confirmed by a Meta spokesperson. However, the strategic direction remains unchanged: Meta will not develop new VR content for Horizon Worlds, and its development focus is exclusively on the mobile version.
Why did Meta decide to shut down Horizon Worlds in VR in the first place?
The shutdown decision was a product of converging pressures. Reality Labs, the Meta division responsible for VR, had accumulated $83.6 billion in cumulative operating losses through 2025. Horizon Worlds never attracted more than a few hundred thousand monthly active users in VR, far below internal targets. Global VR headset shipments had declined for three consecutive years, and Meta’s Quest headset sales specifically dropped 16% year-over-year from 2024 to 2025. The development team was also splitting resources between building for VR and building for mobile, which Meta CTO Andrew Bosworth identified as an efficiency drag. The shutdown announcement was part of a broader restructuring that included laying off approximately 1,500 Reality Labs employees in January 2026 and closing several in-house VR game studios.
How much money has Meta lost on the metaverse and Reality Labs?
Reality Labs has posted an operating loss every year since Meta began disclosing it as a separate segment in 2020. The cumulative operating loss through 2025 reached $83.6 billion. In 2025 alone, the division lost $19.1 billion, up from $17.7 billion in 2024. The single-quarter record was Q4 2025, when Reality Labs posted a $6.02 billion operating loss on $955 million in revenue. Meta’s total company revenue in 2025 was $200.97 billion, meaning Reality Labs generated roughly 1% of company revenue while accounting for losses that would represent a significant standalone business expense by any benchmark.
What did Meta CTO Andrew Bosworth say about the reversal?
Bosworth made his announcement through an Instagram Stories Q&A on March 19, 2026 — one day after the original shutdown announcement. He stated: “We have decided, just today in fact, that we will keep Horizon Worlds working in VR.” He attributed the reversal directly to user feedback, citing “the fans who reached out,” including one user who wrote that they were “heartbroken” about the shutdown. Bosworth confirmed that while VR access would be preserved, Meta’s active development focus for Horizon Worlds would remain on mobile. A Meta spokesperson separately confirmed his statements to TechCrunch.
How many users did Horizon Worlds actually have?
At its peak, Horizon Worlds had fewer than 200,000 monthly active users. This peak was recorded in October 2022, just months after the platform’s wider public launch in late 2021. Internal targets for the platform had been set at 500,000 monthly active users, revised down to 280,000, and the platform never reached even that reduced figure. For reference, competing social VR platform VRChat and mobile platforms like Roblox maintained orders of magnitude more active users without anything close to Meta’s marketing investment or financial backing.
What is happening with the mobile version of Horizon Worlds?
Meta launched a mobile version of Horizon Worlds in September 2023, designed to reach users who do not own VR headsets. As of March 2026, the mobile app had accumulated 45 million total worldwide downloads across iOS and Google Play. Downloads in 2026 were running at approximately 1.5 million, a 53% year-over-year increase. However, total lifetime consumer spending on the app was estimated at approximately $1.1 million — a figure that reflects weak monetization despite the download volume. Meta has stated that all new Horizon Worlds development will focus on the mobile platform.
What happened to Reality Labs employees?
Meta reduced Reality Labs staff by approximately 10% — roughly 1,500 employees — in January 2026. This was the largest single round of cuts in the division’s history, though it followed earlier reductions in 2022, 2023, and mid-2024. The January 2026 cuts specifically targeted teams working on Horizon Worlds content, including Ouro Interactive, an in-house studio created in 2023 to develop first-party VR content. Meta also retired its VR Workrooms app for virtual meetings and paused third-party Horizon OS headset development. Reports as of March 2026 indicate Meta may be considering an additional round of broader company-wide layoffs that could affect up to 20% of total headcount.
Is VR itself failing, or just Meta’s metaverse strategy?
Both elements are present in the data, though they are distinct problems. VR as a consumer hardware category has faced declining shipments for three consecutive years — global headset shipments fell 12% in 2024 and 14% in H1 2025. Meta’s Quest headsets fell 16% in sales from 2024 to 2025. Apple scaled back Vision Pro production due to low demand. These are category-level challenges, not Meta-specific ones. At the same time, Meta’s specific implementation — a social VR platform requiring headset ownership, launched without legs on avatars, never finding meaningful organic engagement — had product-specific problems that contributed independently to its failure. VR as a technology has applications in enterprise, healthcare, training, and gaming that remain more viable than the social metaverse use case.
What is Meta investing in instead of the metaverse?
Meta has shifted its investment emphasis significantly toward artificial intelligence. The company has committed $135 billion in capital expenditure for 2026, with a major portion directed at AI infrastructure. Meta has released the Llama family of open-source large language models, integrated AI assistants across WhatsApp, Instagram, and Facebook, and launched Meta AI as a standalone product. On the hardware side, the Ray-Ban smart glasses product — which runs on AI rather than VR — has emerged as a genuine success, selling over 7 million units in 2025 alone, tripling year-over-year. Zuckerberg has described glasses and AI wearables as the primary hardware focus going forward, building toward devices that provide an ambient AI layer over physical life.
How did Meta’s Ray-Ban smart glasses perform compared to Quest headsets?
The contrast is significant. Meta’s Ray-Ban smart glasses, developed in partnership with EssilorLuxottica, sold over 7 million units in 2025 — tripling sales year-over-year and prompting EssilorLuxottica’s shares to hit an all-time high. Meta also launched Oakley-branded smart glasses through the same partnership. Quest headset sales, by comparison, fell 16% year-over-year from 2024 to 2025, continuing a multi-year decline. The Ray-Ban glasses succeed by integrating into an existing consumer product category (eyewear) rather than asking users to adopt entirely new hardware and usage behaviors. Meta is explicitly directing its wearables investment toward this product line going forward.
Could Horizon Worlds be revived in VR at some point?
Based on available information, this appears unlikely in the near to medium term. Meta has made no commitments to resume VR development for Horizon Worlds. The development team that was building VR content has been restructured. The company’s strategic investment is directed toward mobile Horizon and AI wearables. The VR version survives in maintenance mode — existing worlds accessible to existing users, but without a development pipeline behind it. A meaningful revival would require a significant reversal of current resource allocation, a hardware turnaround in Quest headset sales, and evidence that VR social platforms can attract users at scale — none of which is currently indicated by available data or Meta’s stated plans.
What should brands and marketers take away from the Horizon Worlds story?
Several practical points emerge directly from the data. First, platform audience data deserves rigorous verification before investment — download counts and user registrations can diverge sharply from active, engaged users who are receptive to brand messaging. Second, hardware dependency creates a ceiling on platform reach that marketing budget cannot raise. Third, major technology platforms can shift or discontinue their commitments to specific products quickly and without advance notice to the businesses building on them. Fourth, the most durable question for any emerging platform investment is whether your target customers are actually using it, not whether the platform is conceptually interesting or backed by a major technology company. The Horizon Worlds experience is a case study in what happens when those criteria are not met.
What does this mean for the future of the metaverse as a concept?
The term “metaverse” is being redefined rather than abandoned. Meta’s Bosworth has described it in 2026 as encompassing AR experiences, digital overlays on physical spaces, and even smartphone use in shared physical environments. Under that broader definition, the metaverse becomes a general descriptor for digital-physical integration rather than a specific reference to immersive VR social worlds. Whether that redefinition is meaningful or merely semantic depends on what products eventually get built around it. The immersive social VR version of the metaverse that Meta promised in 2021 — a billion users, hundreds of billions in digital commerce, a replacement for screens — has not materialized and is no longer the company’s focus.
What is the significance of the word “metaverse” disappearing from Meta’s earnings calls?
By the third quarter of 2024, Meta’s executives had stopped using the word “metaverse” in earnings calls entirely. This was a notable departure from prior quarters, when the term had been central to how leadership framed the company’s strategy and justified Reality Labs’ losses to investors. The shift coincided with the post-ChatGPT acceleration of AI investment across the industry. Analysts and observers interpreted the language change as a quiet acknowledgment that the metaverse narrative had not played out as projected and that the company’s attention had moved elsewhere. The language in Q4 2024 and 2025 earnings calls centered on AI engagement, AI-powered advertising, and Llama model performance rather than virtual worlds or social VR.
The Horizon Worlds story is not ultimately about a social media platform failing to attract users, though that is part of it. It is about the distance between a stated vision and what users and markets actually respond to — and about what happens when one of the world’s largest technology companies discovers that distance at a cost of $83.6 billion. The 48-hour reversal on the VR shutdown is, in that context, a minor footnote. Meta kept the lights on for a platform that serves a small, loyal user base, while the development resources and strategic attention moved on. The VR version of Horizon Worlds will remain accessible. It will just exist without any meaningful investment behind it — which, for a platform that spent five years failing to grow, amounts to much the same outcome.
Frequently Overlooked Facts About This Story
For those who have read this far and want the details that most coverage has not emphasized:
The $1.1 million in total lifetime consumer spending on Horizon Worlds mobile — against 45 million downloads — is arguably the single most telling number in this entire story. It suggests that even when Meta successfully acquired user attention on mobile, it could not convert that attention into economic activity at any meaningful rate.
The mobile app’s 53% download growth rate in early 2026 is being cited as evidence of momentum. But comparing absolute scale — 1.5 million downloads in early 2026 versus Roblox’s 65 million daily active users — makes that momentum figure look like what it is: early-stage growth on a small base, not confirmation of a product that has found its audience.
The Ray-Ban glasses story is the actual hardware success. Seven million units in one year is not a niche product. It is a commercially meaningful result in a consumer hardware category. The fact that it runs on AI, not VR, and looks like ordinary sunglasses, is not incidental to its success — it is the explanation for it.
About ALM Corp
ALM Corp is a digital strategy and marketing consultancy that helps businesses make clear-eyed decisions about where to invest their digital presence, content, and media budgets. The Meta Horizon Worlds story is precisely the kind of scenario ALM Corp works with clients to avoid: committing resources to platforms based on projected audience potential rather than verified, current user behavior.
ALM Corp’s services include digital strategy, SEO, content marketing, paid media, and platform evaluation — the combination of capabilities that helps organizations identify which platforms, formats, and channels are actually reaching their customers today, not which ones have been projected to matter in three years. As the technology landscape continues shifting — from social media dominance to AI-native experiences, from VR ambitions to AI wearables, from desktop to mobile to ambient computing — businesses need a strategy partner that reads the available data without agenda and builds plans around what is actually working.
Whether you are reassessing your social media investment after the Meta metaverse retreat, evaluating your content strategy in an era of AI-generated search results, or trying to understand where your brand’s digital audience is actually spending time, ALM Corp brings the analytical rigor and hands-on execution to move from questions to action. The Horizon Worlds story is a reminder that even the largest technology companies can build for years in the wrong direction. The firms that avoid that mistake are the ones that ask the hard questions early.
To learn more about how ALM Corp can support your digital strategy, visit www.almcorp.com.
