Why the Metaverse Disappeared

In October 2021, Mark Zuckerberg stood in front of a camera and announced that Facebook would no longer be called Facebook. The company was now Meta, and its future, he said, belonged to a fully immersive digital world where people would work, socialize, and shop as avatars. Investors cheered. Rivals scrambled to stake their own claims. Roblox, Microsoft, and Nvidia all rushed to attach themselves to the same word.

Four years later, that word had almost vanished from Meta’s own earnings calls. The company’s virtual reality division had lost more money than most countries spend on their space programs, its flagship virtual world had fewer active users than a mid-sized regional newspaper, and Zuckerberg’s public attention had shifted almost entirely to artificial intelligence and smart glasses.

So what actually happened? The metaverse didn’t fail because the idea was absurd on its face. Virtual worlds, avatars, and persistent online spaces already existed and, in some forms, still thrive. It failed because the specific version of the metaverse that Silicon Valley tried to sell the public in 2021 was built on a bet about human behavior that never paid off, propped up by a hype cycle that outran the technology, and eventually abandoned once a more profitable idea came along.

A Word Everyone Suddenly Needed

The term “metaverse” came from Neal Stephenson’s 1992 novel Snow Crash, where it described a shared virtual space people entered to escape a bleak physical world. For decades it stayed a niche reference among game designers and science fiction readers.

That changed in 2021, for reasons that had less to do with any breakthrough in the technology than with timing. The pandemic had pushed enormous numbers of people into video calls, online games, and virtual social spaces almost overnight. Roblox, a platform where users build and inhabit their own game worlds, went public in March 2021 and was suddenly valued in the tens of billions of dollars. Executives at companies with no obvious connection to virtual reality began using the word “metaverse” in earnings calls simply because investors responded to it.

Facebook’s rebrand in October 2021 turned a buzzword into a corporate strategy. Zuckerberg described a fully realized digital layer over daily life, complete with virtual offices, concerts, and storefronts, and he committed Meta’s engineering and marketing muscle to building it. Other companies followed his lead, less because they had a clear plan than because no one wanted to be the one left out of the next internet.

The Product Never Matched the Pitch

Meta’s actual flagship product, a social app called Horizon Worlds, arrived far less finished than the vision it was supposed to embody. Early versions of the avatars had no legs, a design choice made because full-body movement was still too difficult to animate convincingly through a VR headset. The visual quality was closer to a mobile game from a decade earlier than to the richly rendered future Zuckerberg had described.

The gap became measurable, not just anecdotal. Internal Meta documents reviewed by The Wall Street Journal in 2022 showed that Horizon Worlds had reached only around 200,000 monthly active users, far below the company’s original target of 500,000 by the end of that year. Meta had already lowered its own goal once, to 280,000, and was still missing it. NewsBytes

The retention numbers told an even harder story. According to the same documents, most users didn’t return to Horizon Worlds after their first month on the platform, and the total number of users had been declining steadily since spring. Of the more than 10,000 virtual worlds users had built inside the app, fewer than 10 percent were ever visited by as many as 50 people, and most were never visited at all. One internal Meta document summarized the problem in a single blunt line: “an empty world is a sad world.” CNBC + 2

This was the core failure hiding underneath the marketing. A metaverse is only as compelling as the number of other people inside it, and Meta had built the empty stadium before it had convinced anyone to buy a ticket.

Money Poured In, and Almost None Came Back

Enthusiasm might survive a rough beta version of a product. It rarely survives the kind of financial losses Meta was posting. The company organized its virtual and augmented reality work under a division called Reality Labs, and beginning in 2020 it started reporting that division’s results separately, which made the scale of the bet impossible to hide.

The losses grew every single year. Reality Labs lost roughly $6.6 billion in 2020, then $10.2 billion in 2021, $13.7 billion in 2022, $16.1 billion in 2023, $17.7 billion in 2024, and $19.2 billion in 2025, according to Meta’s own financial disclosures. By early 2026, the cumulative total since the division’s creation had passed $80 billion. Revenue, meanwhile, stayed in the hundreds of millions of dollars per quarter — a rounding error next to the losses it was supposed to justify.

For comparison, Meta’s core advertising business generated tens of billions of dollars in profit during the same years. Reality Labs wasn’t a promising startup burning cash to grow; it was a shrinking niche product being subsidized indefinitely by a wildly profitable social media empire. That arrangement can continue for a long time, but it cannot continue forever without investors demanding an explanation.

Investors Ran Out of Patience Before Consumers Ran Out of Novelty

Meta’s stock price absorbed the message before most consumers did. Shares fell roughly 76 percent between their September 2021 peak and their low point in late 2022, a decline that erased hundreds of billions of dollars in market value. Wall Street had tolerated ambitious long-term bets before, but rarely ones with so little revenue to show for so much spending.

Zuckerberg responded in early 2023 by declaring what he publicly called the “Year of Efficiency.” Meta cut more than 21,000 jobs across the company and, crucially, began signaling that resources would shift away from the metaverse’s speculative virtual-world ambitions and toward projects with a nearer-term payoff. The stock recovered sharply once investors saw the belt-tightening take hold, which sent an unmistakable signal about what the market actually wanted from Meta: discipline, not vision.

That signal only grew louder as a new technology arrived to compete for the same investment dollars. Generative artificial intelligence exploded into public awareness in late 2022 and 2023, giving Meta, its competitors, and its investors a fresh and far more commercially convincing story to tell. Advertising dollars, engineering talent, and boardroom attention all began flowing toward AI systems that could plausibly improve Meta’s existing multibillion-dollar ad business within a single product cycle, rather than toward a virtual world that might or might not exist a decade from now.

The Hardware Problem Nobody Solved

Even a perfectly designed virtual world needs a way for people to enter it, and virtual reality headsets never became something most people wanted to put on their heads. Meta’s Quest headsets sold in the millions, which sounds impressive until it’s measured against the billions of smartphones already in use worldwide. Headsets were bulky, could cause motion sickness during extended use, and typically sat in a closet after a few weeks of novelty.

Meta’s own internal data reflected this pattern. Retention rates for Quest headsets dropped in each of several consecutive years, with more than half of buyers no longer using the device within six months of purchase. A product needs sustained daily use to anchor an entire alternate digital economy, and headsets were instead behaving like exercise equipment: purchased with enthusiasm, then abandoned.

This wasn’t a failure of imagination so much as a failure of physical comfort and clear purpose. People will tolerate an imperfect app on a device they already carry everywhere. They are far less forgiving of a device that has to be strapped to the face before the imperfect app can even load.

What Popular Memory Gets Wrong

It’s tempting to conclude that the metaverse simply died, but that framing oversimplifies what actually happened. Meta did not shut down Reality Labs. As late as early 2026, the division was still shipping products, still investing billions of dollars annually, and still holding a majority share of global VR headset shipments.

What changed was the story being told about that spending. Meta began describing its hardware ambitions primarily in terms of smart glasses and AI-powered wearables rather than fully immersive virtual worlds, a category where consumer interest and Meta’s own commercial results had proven considerably stronger. Reports in late 2025 indicated Meta was cutting Reality Labs’ internal budget by as much as 30 percent while shifting the freed-up spending toward AI infrastructure, a move investors welcomed as a long-overdue correction rather than an abandonment of the underlying hardware business.

In other words, the ambition to build immersive alternate worlds didn’t disappear so much as it lost its argument for being the company’s central identity. The word “metaverse” became a liability to say out loud, even as some of the underlying technology quietly continued to develop under less grandiose names.

The Lesson Behind the Collapse

The metaverse’s rise and retreat says less about virtual reality as a technology than about how quickly a corporate narrative can outrun the product supporting it. A headline-grabbing rebrand, a wave of copycat investment, and a hype cycle fed by pandemic-era isolation combined to convince an entire industry that a specific vision of digital life was inevitable, before anyone had actually built something people wanted to use every day.

The technology that survived the collapse of the metaverse hype wasn’t the immersive virtual office or the legless avatar meeting room. It was the far more modest smart glasses and AI assistants that fit into a life people already had, rather than asking them to build an entirely new one. That distinction may be the most durable insight from the whole episode: transformative technology rarely arrives by asking people to abandon the world they know for one that only exists in a pitch deck.

Frequently Asked Questions

Did Meta completely shut down its metaverse projects?

No. Reality Labs continued operating and investing billions of dollars annually as of early 2026, though the company sharply reduced its public emphasis on immersive virtual worlds and shifted investment toward AI-powered smart glasses and wearables instead.

How much money did Meta lose on the metaverse?

Meta’s Reality Labs division, which houses its virtual and augmented reality efforts, accumulated more than $80 billion in operating losses between 2020 and early 2026, with annual losses growing nearly every year during that period.

Was Horizon Worlds the only reason the metaverse lost momentum?

No. Weak user retention in apps like Horizon Worlds was one factor, but uncomfortable and expensive VR headsets, mounting financial losses, and the sudden rise of generative AI as a more commercially convincing investment story all contributed to the shift.