Facebook succeeded by connecting people to things they already wanted. The metaverse failed by asking people to want something new. That difference — simple in description, devastating in consequence — explains the $80 billion gap between Meta’s two defining products. Horizon Worlds is being shut down on VR, removed from the Quest store in March and terminated on June 15, after close to $80 billion in losses. Mark Zuckerberg understood what made Facebook work. He did not apply that understanding to the metaverse.
Facebook succeeded because it connected people to a desire they already had — the desire to share their lives with friends and know what was happening in the lives of people they cared about. The product satisfied a pre-existing motivation so effectively that it spread organically, without requiring users to develop new habits or desires. The viral growth that made Facebook valuable was driven by genuine human need, not product engineering.
The metaverse asked for something different. It asked users to develop a new desire — the desire to spend meaningful time in virtual environments, interacting with avatar representations of people and participating in digital economies. That desire existed in some communities, particularly in gaming. But it was not a mainstream motivation, and Horizon Worlds could not generate it through product features or marketing investment.
Reality Labs spent close to $80 billion in the attempt. Layoffs of more than 1,000 employees in early 2025 confirmed that the attempt had been unsuccessful. Meta’s pivot to AI reflects, consciously or not, a return to the Facebook model — building tools that satisfy existing desires rather than products that require new ones. AI makes people better at things they already want to do. The metaverse asked them to want to do new things.
The lesson from the gap between Facebook and the metaverse is as old as the technology industry: find a genuine need and satisfy it better than anyone else. Zuckerberg did it once, spectacularly. The metaverse demonstrated that even the most successful practitioners of the formula can forget it. The AI era gives him the opportunity to rediscover it.