Meta reportedly burned through $80 billion in pursuit of a 1992 vision that its founder, author Neal Stephenson, now describes as fundamentally flawed. The man who coined the term "metaverse" in his cyberpunk novel Snow Crash has become the vision's most unlikely critic.
Stephenson argues that glasses-based displays create a privacy and trust problem. "When someone holds up their phone and aims it at you, it's obvious that you are on camera. That's not true in the case of glasses or goggles," he wrote. "So it's creepy." For investors watching Meta hemorrhage capital, Stephenson's reversal offers validation of a market verdict that took five years and tens of billions of dollars to confirm.
The numbers speak for themselves. Meta's Reality Labs division posted $19.2 billion in losses in 2025 alone, bringing cumulative metaverse spending past $80 billion since 2020 with no viable consumer product to show for it. Horizon Worlds peaked at roughly 300,000 monthly active users in early 2022, a fraction of the 500,000 target Meta set for that year and nowhere near the billion-user vision Zuckerberg promised.
The fiscal responsibility argument against the metaverse is straightforward: unlimited capital cannot substitute for product-market fit. Stephenson has acknowledged his original reasoning was contextual. In 1990, when writing Snow Crash, computer graphics emerged through massive, heavy CRTs with terrible resolution. Head-mounted displays seemed like the logical next step. But technology evolved differently.
Instead, gaming platforms like Roblox now host something like 380 million monthly active users accessing immersive 3D worlds through flat screens on smartphones and computers. The metaverse came to market without needing VR headsets. It simply arrived through existing technology that consumers already trusted and used.
The counterargument, which Meta might still make, centres on sunk costs and transitional value. A lot of the VR research Meta funded, including computer vision, environmental mapping, sensors, and optics, also happens to be the core infrastructure needed for practical augmented reality, which may have functioned as a stepping stone toward mixed reality systems rather than a pure attempt to build the fully virtual Metaverse. This perspective holds that the spending wasn't entirely wasted, even if the original vision failed. That argument carries some weight, but it doesn't justify the scale of commitment relative to results.
What stands out is how institutional inertia and executive ambition can override empirical evidence. The platform never achieved product-market fit, with retention consistently poor even among the handful of users who adopted it. Yet Meta continued pouring capital into Reality Labs for years, subsidising a product that consumers had already rejected. That pattern of behaviour, repeated across technology companies, suggests a deeper problem: senior leaders rewarded for bold bets rather than results.
The pivot to artificial intelligence, while sensible, shouldn't obscure what went wrong. The company is now redirecting resources toward AI infrastructure, with 2026 capital expenditures projected to reach $135 billion for data centres and AI compute. This reallocation indicates management finally recognising the metaverse as a failure. But that clarity took far longer than it should have, and the cost was staggering.
For taxpayers concerned about tech billionaires squandering resources on vanity projects, this episode offers uncomfortable lessons. In fourth-quarter earnings posted in January, Meta's Reality Labs unit reported an operating loss of $6.02 billion in a single quarter. That's a public company with shareholder obligations, not a research lab. The pressure on quarterly earnings suggests the metaverse was never a reasonable long-term bet; it was a rebranding gambit that got away from its creators.
Stephenson's recent writing suggests he views the situation with wry detachment. He describes being "curiously detached" from Meta's metaverse project, receiving a text from a former colleague reading "Sorry for your loss" when Facebook changed its name. In retrospect, that message was prescient, marking the moment when the Metaverse became his "alienated, prodigal brainchild."
The genuine complexity here is worth acknowledging. Technological futures are difficult to forecast. First-mover advantage in immersive computing might have had real value five or ten years from now. But strategy requires balancing conviction against feedback, and Meta ignored both consumer behaviour and the voice of its own visionary. For a company of Meta's resources, that choice was defensible only as long as results remained hidden. The platform never drew more than a few hundred thousand monthly active users, a fraction of what would have been needed to justify the investment.
The lesson isn't that ambitious technology bets always fail. The lesson is that they must be tested ruthlessly against reality, and killed quickly when evidence mounts. Meta did the opposite. It spent five years subsidising a product nobody wanted, defended a vision nobody outside the executive suite believed in, and rewarded employees for building features for an audience that wasn't coming. Stephenson's about-face simply made official what the market had already decided.