Meta to Invest Up to $145 Billion in AI Infrastructure by 2026

Meta Platforms (NASDAQ: META) has revised its capital expenditure (capex) forecast for 2026, announcing an anticipated spend of between $130 billion and $145 billion on artificial intelligence (AI) infrastructure. This decision, which saw the lower end of its previous guidance raised, underscores the company's aggressive pursuit of AI capabilities and its commitment to substantial investments in this transformative technology. The upward adjustment reflects Meta's strategic focus on AI as a core driver for its future growth.

Initially, Meta's capex guidance for 2026 was set between $115 billion and $135 billion at the beginning of the year. Following its first-quarter results, this range was increased to $125 billion to $145 billion. The latest revision solidifies Meta's determination to expand its data center and hardware capacity for training advanced AI models, enhancing existing products, and developing new AI-driven businesses. CEO Mark Zuckerberg has articulated that these investments are aimed not only at improving current offerings but also at building entirely new AI-driven ventures. Key aspects of this strategy include ongoing investment in the Llama family of open-source AI models and the development of new products like personal AI agents.

However, these large-scale investments have significantly impacted Meta's financial performance. The company's free cash flow (FCF) for the second quarter of 2026 plummeted by 91% year-over-year to just $784 million. This represents Meta's weakest FCF performance since late 2022, highlighting the mounting financial pressure from its aggressive AI infrastructure buildout. Concurrently, operating income declined by 8% in the same quarter, while costs and expenses surged by 55%. The market reacted negatively to these escalating costs and the sharp drop in FCF, causing Meta's shares to fall by approximately 10% in extended trading following the second-quarter earnings report.

Meta's substantial spending is part of an unprecedented AI infrastructure race among major technology companies. Across the industry, Big Tech firms are projected to invest over $700 billion in AI-related infrastructure through 2026. This trend indicates that companies are taking significant risks to avoid falling behind in the competitive AI landscape. For instance, while Microsoft (NASDAQ: MSFT) also increased its AI investments, strong Azure cloud growth led to a positive investor response. In contrast, Alphabet (NASDAQ: GOOG) reported its first-ever quarter of negative free cash flow, underscoring the immense costs associated with this race.

Analysts and market observers hold varying views on the long-term returns of Meta's aggressive AI strategy. Mark Zuckerberg believes that personal AI agents will become a major consumer technology category and that Meta's vast user base provides a unique opportunity to commercialize AI products at scale. Some analysts consider the stock's sell-off an overreaction, pointing to Meta's robust advertising business as a funding source for these investments and the potential for new revenue streams by leasing surplus computing capacity to other companies in the future. Nevertheless, the timeline for these investments to translate into tangible revenue streams and the extent of investor patience remain key uncertainties.