March 24, 2026
We Were All Supposed to Live in the Metaverse by Now
Few technology failures have been quite as public, or quite as expensive, as the metaverse. It began as a grand declaration about the future of human connection, ended this month with a partially reversed shutdown announcement, a workforce reduced by thousands, and a pivot so complete that the company that renamed itself after the concept has stopped using the word entirely. This is how it happened and what it means for the next big thing.
October 2021: The Announcement
In October 2021, Facebook was renamed Meta Platforms, with chairman Mark Zuckerberg declaring a company commitment to developing a metaverse. The term itself had been coined in a 1992 science fiction novel, but for most people, this was the moment they first heard it. Zuckerberg called it “the next frontier” and wrote: “Our hope is that within the next decade, the metaverse will reach a billion people, host hundreds of billions of dollars of digital commerce, and support jobs for millions of creators and developers.”
The announcement prompted enthusiasts to declare the metaverse as the world’s new computing interface. Bill Gates added to the chorus, predicting at the time that within two to three years, most work meetings would shift from video screens into virtual metaverse environments. Analysts matched the optimism, with Gartner predicting that 25% of people would spend at least one hour per day in the metaverse by 2026, and that 30% of companies worldwide would have metaverse-ready products and services by that date. The word “metaverse” came runner-up in the Oxford English Dictionary’s public vote for word of the year.
Meta announced it would invest $10 billion and hire 10,000 people in Europe to build out a metaverse. Reality Labs, its division dedicated to VR hardware, software, and content, became the engine of that ambition.
2020–2021: How the Pandemic Fuelled the Hype
COVID-19 had spent the previous 18 months forcing millions of workers into remote set-ups, and the limitations of flat video calls were becoming a cultural reference point. The effects of Zoom fatigue and awkward attempts to recreate informal office interactions were partly behind the growing interest in the metaverse as a potential future of remote work.
Virtual office and immersive technology businesses saw a wave of major investment. Start-up Gather raised $77 million, Teamflow raised $50 million, and longer-standing virtual office company Virbela saw its revenues increase by 260% in the second quarter of 2020 alone.
During this period, some enterprises began exploring VR as a training tool. Verizon, for example, began running VR-based training for contact centre agents in 2021, using immersive scenarios to prepare staff for high-pressure customer interactions.
2022: The Cracks Appear
Horizon Worlds launched to the public in early 2022, and the reception was poor from the start. By October 2022, the platform had fewer than 200,000 monthly active users, well below an internal target of 280,000, which had itself been revised down from an original goal of 500,000. Player avatars launched without legs, and negative feedback set an early tone that the platform never fully recovered from.
Meta’s Reality Labs posted a $13.7 billion operating loss for 2022, Microsoft cut staff from its Mixed Reality and HoloLens teams, and consumers returning to in-person life showed little appetite for VR headsets.
The customer-facing metaverse that CX analysts had predicted, like VR storefronts, avatar-based service agents, and immersive brand experiences, was not going to happen. Although Nike built Nikeland on Roblox, Samsung created a virtual replica of its New York flagship on Decentraland, and luxury retailers experimented with AR try-ons, these were marketing exercises as much as genuine service propositions. For the most part, customers did not follow brands into virtual worlds with any enthusiasm.
2023: Reality Catches Up
Reality Labs losses increased to $16.1 billion in 2023. Disney cut its metaverse division, and media reports proclaimed the metaverse was dead. In February 2023, Zuckerberg announced the formation of a new product group dedicated to generative AI, consolidating teams that had previously been scattered across the company.
Trading volume in platforms like Decentraland and The Sandbox dropped over 90% from their 2021–2022 peaks. The virtual land that companies had paid millions to acquire had become essentially worthless. In its 2024 debut of the Apple Vision Pro headset, Apple took pains to disassociate the device from the metaverse entirely, calling it an entrée to “spatial computing” instead. By Q3 2024, Meta executives had stopped using the word “metaverse” on earnings calls altogether.
March 2026: The End
Reality Labs burned through more than $80 billion since 2021. Quest headset sales fell 16% year-on-year between 2024 and 2025, according to IDC. In January 2026, Meta made significant cuts in its Reality Labs division, impacting over 1,500 employees and shuttering several game studios.
Then, on 18 March 2026, Meta sent an email to Horizon Worlds users telling them their virtual reality social platform was being shut down. By June 15, the app would only work on mobile. The announcement landed with relatively little public mourning, which itself said something about how far the platform had fallen from its original ambitions.
Within 48 hours, Meta partially reversed the decision after user pushback. CTO Andrew Bosworth said the company had decided to keep Horizon Worlds running in VR “for the foreseeable future,” though with limited support and mainly for people using existing games. No new development will resume.
The most durable legacy of the metaverse era for the CX industry is not what happened in the backend. Enterprises that deployed VR as a contact centre training tool found genuine results. Better-prepared agents handled difficult interactions more effectively, and evidence suggested they were less likely to leave. AI is now doing what the metaverse was supposed to do, delivering personalised, scalable customer interactions without requiring anyone to strap on a headset.
Could AI Be Next?
The metaverse rose and fell in a way that will be familiar to anyone who has watched a technology hype cycle play out before. Outsized ambition met an unexpected accelerant in the pandemic, investment ran far ahead of user demand, and the correction was as swift as the enthusiasm had been. It is a reasonable question to ask whether AI is following the same arc.
AI is already embedded in products that hundreds of millions of people use daily, through search engines, writing tools, customer service interfaces, in ways that Horizon Worlds never was. Adoption across the CX and contact centre industry alone has gone from 62% to 80% in a single year, and the productivity gains are already being measured. Where the metaverse required users to change their behaviour dramatically, like buy hardware, AI meets people where they already are.
That said, some of the warning signs are familiar. Analyst predictions have become increasingly extravagant, while enterprise AI projects are running into execution problems at scale. The lesson the metaverse offers is not that transformative technology always fails, but that hype cycles are expensive, and the platforms that survive are the ones that solve a problem users have, not ones they are supposed to develop.
