Meta to Cut 8,000 Jobs While Spending $125-145B on AI Infrastructure, Reorganizing Into 5-12 Person 'AI Pods'
Tags AI · Infrastructure · Enterprise
Meta will eliminate 8,000 positions (~10% of its 77,000 workforce) on May 20, 2026, while cancelling 6,000 open requisitions, for an effective headcount reduction of 14,000. Simultaneously, Meta is increasing capital expenditure to a record $125-145 billion for AI infrastructure. CEO Mark Zuckerberg is reorganizing surviving teams into 5-12 person 'AI pods' under Chief AI Officer Alexandr Wang's Superintelligence Labs, with some teams operating at a 1:50 manager-to-engineer ratio. CFO Susan Li admitted she doesn't know what Meta's ideal headcount looks like given rapidly evolving AI capabilities. Meta stock dropped 6% after the earnings call.
Technical significance
Meta's restructuring is the most aggressive example yet of a major tech company reorganizing around AI-native team structures. The $125-145B capex figure — nearly double the prior year — shows the scale of the bet that AI infrastructure investment can offset headcount reduction. The 'AI pod' model (5-12 person teams owning one production capability end-to-end) could become a template for other large tech companies if it proves effective.