Every company has a finite capacity for expensive mistakes. Meta is shutting down Horizon Worlds on VR — removed from the Quest store in March, fully terminated on June 15 — after close to $80 billion in losses. Mark Zuckerberg’s metaverse has consumed an extraordinary share of that capacity. The real question is whether Meta — financially robust but reputationally challenged — can afford another mistake of this magnitude as it commits to AI.
The financial capacity is less in question than the reputational and organizational capacity. Meta’s core advertising business generates tens of billions in annual revenue, providing a financial cushion that makes even $80 billion in losses survivable. The company is not in financial distress from the metaverse failure; it is, however, in a different kind of distress — the kind that comes from sustained strategic failure in a highly visible domain.
The reputational capacity concern is real. Investors, employees, regulators, and the public have all updated their assessments of Zuckerberg’s strategic judgment based on the metaverse. Each of those stakeholder groups now approaches Meta’s AI commitments with more skepticism than they brought to the metaverse in 2021. Rebuilding confidence requires demonstrating that the AI strategy is better grounded in evidence and more likely to generate commercial returns.
The organizational capacity concern involves talent and culture. Layoffs of more than 1,000 Reality Labs employees in early 2025 affect the company’s ability to attract and retain the people it needs for sustained technological investment. If talented engineers and researchers observe that Meta’s strategic pivots carry personal career risk, the company’s ability to recruit at the frontier of AI development may be compromised.
The AI era thus begins with Meta’s capacity for error already partly consumed by the metaverse. A second major strategic failure — an AI bet that generates comparable losses without commercial returns — would be qualitatively different from the first. It would suggest a systematic problem with strategic judgment rather than a single expensive miscalculation. Zuckerberg and Meta need to get AI substantially right. The capacity for another $80 billion mistake is not guaranteed.