Some products collapse under the weight of what they were promised to be. Meta has shut down Horizon Worlds on VR — off the Quest store by March, terminated on June 15 — after close to $80 billion in losses. Mark Zuckerberg promised the metaverse would reach a billion users, host hundreds of billions in commerce, and support millions of creative jobs. The product that was built could not carry that weight. Close to $80 billion confirms the gap between the promised weight and the product’s actual capacity.
The promises were specific and public. The billion-user figure, the hundreds-of-billions in virtual commerce, the millions of creator jobs — each was articulated in Zuckerberg’s 2021 blog post as the expected outcome of the metaverse investment. Each was presented not as aspirational but as expected — the projected returns on a long-term but ultimately achievable bet. The specificity of the promises created specific accountability for the outcomes.
Horizon Worlds could not bear the accountability. Monthly active users in the hundreds of thousands against a billion-user projection represented a failure by orders of magnitude. Virtual commerce that never developed at meaningful scale against hundreds-of-billions projections represented a fundamental commercial failure. The promises established a standard that the product could not approach, let alone meet.
Reality Labs spent close to $80 billion trying to build a product capable of meeting the standard the promises established. The investment improved the product; the gap between the promises and the product never closed. Layoffs of more than 1,000 Reality Labs employees in early 2025 and the formal AI pivot acknowledged that the gap was permanent for the current form of the platform.
The promises have become part of the metaverse’s historical record — evidence of the distance between visionary projection and commercial outcome that the technology industry should study carefully. Zuckerberg’s AI projections, made in the context of the metaverse failure, will be calibrated more carefully by observers who remember what the metaverse promises looked like and how they resolved. The weight of the AI promises must be proportional to what the AI products can actually carry.