Meta to Invest $600 Billion in U.S. AI Infrastructure

On 7 November 2025, Meta Platforms announced that it will commit more than US $600 billion by 2028 to expand AI infrastructure, including building large-scale data centres across the United States and supporting workforce growth.

What the investment covers

Meta said the investment will fund AI data centres, compute infrastructure and workforce expansion, stating the goal is to “support AI technology, infrastructure and workforce expansion” in the U.S.
The company also emphasises community and sustainability commitments: it reports investing in grid upgrades, recruiting local labor, and targeting water-positive and energy-efficient operations in its data centres.
Meta told investors on its Q3 2025 earnings call that “capital-expenditure growth will be notably larger in 2026 than 2025” as AI becomes a primary driver of spending.

Why this matters

The scale of the commitment positions Meta among the rare U.S. tech giants promising infrastructure investment on the order of hundreds of billions of dollars for AI.
AI infrastructure is rapidly becoming a strategic asset: data centres, power, cooling and networking are foundational for training and running large AI models — Meta’s pledge signals its intent to lead in that infrastructure sweep.
The investment has implications for jobs and regional economies: Meta reports that its existing U.S. data-centre projects have supported more than 30,000 skilled-trade jobs and 5,000 operational jobs, and that subcontractor business exceeds US $20 billion across the country.

Risks and scrutiny

Analysts question how Meta will finance such a large investment: with current profits, cash-flow and capital-expenditure levels, sustaining hundreds-of-billions in spending will require either heavy debt, leveraged investments or long pay-back periods.
Meta must deliver returns on its AI-infrastructure build-out: investors are increasingly sensitive to large scale-ups that don’t generate proportional revenue or growth.

What happens next

Meta will likely announce specific projects, locations and partnerships for new data centres as part of its build-out over the next three years.
Observers will watch power-grid implications, regional job creation, sustainability credentials (energy use, water use, materials) and how Meta monetises its infrastructure via AI services.
The announcement raises competition dynamics: as Meta builds out, rivals such as Google, Microsoft and Amazon may accelerate their own infrastructure plans to maintain competitive parity.

In summary: Meta’s pledged $600 billion investment in U.S. AI infrastructure by 2028 marks a dramatic escalation in the tech industry’s infrastructure race. With data centres, compute and talent as the backbone of future AI capabilities, Meta is staking its future on being at the foundation of the next wave of artificial intelligence — but also accepting significant financial and operational risk in doing so.