META Raises 2026 Capex to Record $145B While Cutting 8,000 Jobs in AI Pivot

Meta announced a dramatic workforce restructuring that underscores its commitment to artificial intelligence infrastructure despite broader economic uncertainty. The company raised full-year 2026 capital expenditure guidanceCompany-issued forecasts of future financial performance. to $145 billion, a record high, while simultaneously announcing the elimination of 8,000 jobs, approximately 10 percent of its workforce. Additionally, Meta is canceling 6,000 open roles and reassigning 7,000 employees into AI-focused positions. This dual move, far from austerity, represents a workforce reset designed to maximize returns on AI capex and align headcount with strategic priorities.

Meta's capex acceleration reflects hyperscaler competition intensifying around large language models and generative AI inference. Microsoft, Google, and Amazon each raised AI capex guidanceCompany-issued forecasts of future financial performance. in recent weeks, signaling that competitive AI dynamics are driving infrastructure spending at a pace not seen since the peak of cloud build-outs. Meta's $145 billion guidance suggests the company is committing to large-scale training clusters, inference infrastructure, and data-center build-outs to compete with OpenAI, Anthropic, and in-house AI development efforts. The job cuts allow Meta to redeploy capital toward compute and away from legacy operations.

Investors initially cheered the announcement, viewing the layoffs as necessary cost discipline while the capex increase demonstrated conviction in the AI thesis. Yet skeptics noted that if capex yields diminishing returns or if enterprise AI adoption slows, Meta could face pressure to reduce guidanceCompany-issued forecasts of future financial performance.. The company's metaverse investment tab remains a concern; executives have signaled that Reality Labs capex will eventually moderate, but near-term guidance does not distinguish between metaverse and generative AI spending.

The broader implication: hyperscalers are willing to absorb short-term margin pressure and workforce disruption to secure leadership positions in AI infrastructure. If capex spending translates into defensible competitive moats in LLM training and inference, the investment thesis holds. If capex yields commoditized commodity AI services available through multiple vendors, returns will compress. For now, Meta's willingness to raise capex despite rate hike expectations and geopolitical shocks signals confidence in secular AI demand.