An autopsy of the Meta metaverse reveals a mix of genuine insight and critical miscalculation. Horizon Worlds is being shut down on VR — off the Quest store in March, fully gone from VR by June 15 — after close to $80 billion in losses. Mark Zuckerberg’s vision was not wrong in every dimension; its implementation, timing, and target market were the primary sources of failure.
The insight at the core of the vision — that computing was moving toward more immersive forms of interaction — remains plausible. Every preceding platform shift involved moving toward richer, more expressive interaction: from text to graphical interfaces, from graphical interfaces to touch, from touch toward spatial computing. Zuckerberg’s belief that spatial computing through VR was next was informed by that pattern.
The miscalculation was about the speed of the transition. Horizon Worlds required mass VR adoption to succeed, and VR adoption was not moving fast enough to support a social platform at scale. The platform’s few hundred thousand monthly users were a ceiling imposed by the limited installed base of VR headsets — a ceiling that sustained investment in Horizon Worlds could not break.
Reality Labs lost close to $80 billion attempting to break it. Layoffs of more than 1,000 employees in early 2025 and a pivot toward AI marked the formal end of the attempt. What comes next — AI and wearable devices — reflects a different theory about how spatial computing will achieve mass adoption: not through dedicated VR headsets but through lightweight AR glasses integrated into everyday life.
The AI chapter that follows the metaverse chapter may prove more successful precisely because it does not require a behavioral shift as dramatic as VR demanded. AI integrates into existing behaviors; VR required replacing them. If Zuckerberg has internalized that distinction, the metaverse’s $80 billion may have bought an understanding that shapes a more successful future.