“I don’t even understand what it [metaverse] stands for,” Zoho chief scientist Sridhar Vembu remarked in 2022, just a year after Facebook rebranded itself as Meta, formally launching the metaverse as a corporate strategy rather than merely a cultural idea.
The metaverse—a virtual, shared digital universe where people can interact, work, play, and create using avatars, often blending gaming, social media, and digital commerce—captivated the world in 2021.
Nearly five years on, Meta’s bet on virtual reality came to an end this month, with the company reportedly laying off nearly 1,500 employees from its virtual reality research arm Reality Labs and shutting down several VR game studios, according to The Wall Street Journal. By late 2025, Meta had also cut the metaverse division budget by up to 30%, after incurring over $70 billion in cumulative losses since 2020.
The metaverse, and the promise it held, seems to be a distant memory now.
For Meta, the strategy was a bet on Gen Z’s preference for socialising in online games such as Fortnite and Roblox, rather than traditional social media apps. The pivot also helped the company distance itself from the baggage of the Facebook brand. The pandemic contributed as well, with the rise of working from home; people perceived a need for a more immersive environment, closer to real life, rather than just gaming.
Meta envisaged the metaverse as the next major social platform, where users would connect in a virtual world via Horizon Worlds and play games on VR headsets. Fast-forward, and the metaverse has effectively been abandoned in favour of the euphoria surrounding artificial intelligence.
What Went Wrong
The metaverse was overhyped by investors and analysts alike. According to McKinsey’s Value Creation in the Metaverse report, the technology could generate up to $5 trillion by 2030. Yet early versions were poor products: the software platforms weren’t ready, the hardware lagged—as seen with Oculus Quest and Apple’s Vision Pro virtual reality headsets—and the ecosystem was still evolving.
Meta overpromised a future while its product under-delivered. It was a failure of the “build in the open” model, where early tech products are shipped to consumers in the hope of feedback that drives iteration.
The metaverse isn’t a standalone technology. Its growth depended on the convergence of several building blocks: the access layer (virtual worlds and external reality), the middle layer (Web3, digital twins, AI), and the infrastructure layer (cloud storage, computing, hyperconnected networks). Most failures occurred in the middle layer—Decentraland, The Sandbox, Cryptovoxels, Somnium Space, AI avatars, non-player characters, and digital twin experiments.
Jaspreet Bindra, founder of AI literacy initiative AI&Beyond, describes the metaverse as “a marketing term wrapped around AR, VR, extended reality, and gaming.” He explains in a conversation with AIM that for Mark Zuckerberg, the metaverse was a way to bypass Apple and Google’s 30% app store commissions, and gain control over his ecosystem. While ambitious, it was also costly. In Q3 2025 alone, Reality Labs posted an operating loss of $4.4 billion.
However, Bindra does not believe Meta over-invested early. “In tech, you have to invest early,” he says. “The same is true for AI, with Big Tech pouring billions into the technology.”
Instead, poor consumer adoption, expensive and uncomfortable hardware, and the absence of compelling everyday use cases led to failure. The metaverse aimed to recreate real places virtually, but most users visited once and never returned. Limited interaction and fresh experiences meant only gaming retained engagement.
Lack of safety features further dampened enthusiasm. Meta only introduced its “personal boundary” feature in Horizon Worlds in 2022 after reports of sexual harassment of avatars surfaced, later adjusting it so it was active only with “non-friends,” and allowing users to switch it off entirely.
Metaverse Failures in India
Metaverse-focused initiatives in India had similar setbacks.
Murali Reddy, founder of Bullieverse, a metaverse and Web3 startup, explains how his company had to pivot to agentic AI to sustain its gaming platform. “There was no hardware or infrastructure support. Building basic infrastructure—VR/AR hardware, powerful GPUs, cloud computing, 3D content creation, real-time networking, and data centres—was extremely costly. For most firms, costs rose faster than user adoption.”
Bullieverse failed due to two main issues. During the NFT boom, it sought to bring NFT characters into a 3D metaverse for gaming and interaction. While the idea was sound, high costs and an unsustainable play-to-earn model rendered it unviable.
Similarly, Manish Maheshwari, former head of Twitter India, faced an early setback with Invact Metaversity, a Bengaluru startup creating an immersive metaverse-powered education platform. Within three months of raising a seed round, he left due to internal differences. Maheshwari acknowledged that the vision for Metaversity did not materialise, citing technological and ecosystem limitations.
Meta also struggled to convince investors of the metaverse’s potential. Many were sceptical of the vision for immersive digital spaces and the demand for VR headsets, which underpin the technology. Despite dominating the VR market with Oculus headsets, sales declined. Counterpoint Research reported that global VR headset shipments fell 12% year-on-year in 2024, marking the third consecutive year of decline. Meta accounted for 77% of those shipments.
The metaverse debacle claimed many casualties. Armature Studio, Twisted Pixel, and Sanzaru—part of Meta’s VR and Reality Labs division—were closed. VR fitness app Supernatural, acquired by Meta in 2023, is now in “maintenance mode.” Meta’s VR workplace programme, Workrooms, is also shutting down.
Shift to AI
“The metaverse likely distracted Mark Zuckerberg from AI,” says Bindra. While Meta invested in open-source AI like LLaMA and its research labs, attention was divided. Eventually, Meta refocused on AI with projects like the AI training supercluster Prometheus.
Other Big Tech companies also re-evaluated. Microsoft, after spending $69 billion on Activision Blizzard to bolster gaming, scaled back industrial metaverse projects like Mesh. Google deprioritised AR/VR and Web3 ventures to focus on generative AI. Apple, leveraging a decade of AR research, launched the Vision Pro but rebranded it as “spatial computer”, avoiding the broader metaverse label.
Meta is now cutting metaverse spending, reorganising teams, and investing heavily in AI talent and infrastructure. The company is transitioning from open-source models to proprietary generative AI models code-named Avocado (text and coding) and Mango (multimodal) and reportedly plans to integrate them across Facebook, Instagram, and WhatsApp this year. This pivot signals that AI, not virtual worlds, is now Meta’s primary growth engine.
Bulliverse has similarly shifted to AI. “This is no longer hype—it’s already happening. People actively use AI coding tools. I use them myself; without them, we would have needed three or four more developers. That demonstrates that the technology works, unlike the metaverse,” Reddy notes.
The term ‘metaverse’ was coined by Neal Stephenson in his 1992 science-fiction novel Snow Crash. While Big Tech tried and tried again to bring that fiction into reality, technology and consumer behaviour became its Achilles’ heel.
- Anthropic Makes $2.08 in Revenue for Every $1 Spent on Compute — Here’s the Full IPO Story
- OpenAI Says AGI is Here. But Not Everyone is Buying It
- Why Oracle is Bringing Post-Quantum Security to Java
- What Really Happens When You Turn Your Selfie Into a 1980s AI Pic?
- AI Agents Are Making Data Security A Bigger Enterprise Problem
- OpenAI is Getting Nervous About Reinforcement Learning