NextFin News - Meta Platforms shares tumbled in after-hours trading on Wednesday after the social media giant significantly raised its capital expenditure forecast for 2026, signaling that the cost of winning the artificial intelligence arms race is climbing faster than many investors had anticipated. According to Bloomberg, the company now expects full-year capital spending to land between $115 billion and $135 billion, a massive leap from the $72 billion spent in 2025. This aggressive upward revision overshadowed a first-quarter revenue beat, as the market pivoted its focus from current advertising strength to the daunting long-term bill for AI infrastructure.
The spending surge is primarily driven by the acquisition of advanced hardware, including Nvidia’s Blackwell and Rubin GPUs, alongside the development of proprietary data centers. Mark Zuckerberg, Meta’s Chief Executive, defended the strategy by arguing that the company must build capacity ahead of demand to avoid being sidelined in the generative AI era. However, the scale of the investment—nearly doubling in a single year—has reignited concerns about the "Year of Efficiency" being a distant memory. While revenue for the first quarter reached approximately $55.5 billion, representing robust growth in the core advertising business, the sheer magnitude of the projected cash outflow suggests that profit margins will face sustained pressure throughout the next fiscal year.
Gene Munster, managing partner at Deepwater Asset Management, noted that while Meta is correctly identifying AI as a generational shift, the "sticker shock" of $130 billion in annual spending is difficult for the public markets to digest in a single session. Munster, who has historically maintained a constructive view on large-cap tech’s infrastructure cycles, suggested that this level of spending indicates Meta is seeing internal signals of AI utility that have yet to be fully reflected in external product monetization. His view, however, is not a universal consensus; several sell-side analysts have expressed skepticism, questioning whether the return on investment for these massive server farms will materialize before the current hardware cycle becomes obsolete.
The divergence in market sentiment highlights a growing tension between Silicon Valley’s long-term technical ambitions and Wall Street’s quarterly performance requirements. Meta’s Reality Labs division continues to operate at a significant loss, and the addition of a hundred-billion-dollar AI infrastructure bill leaves little room for error in the company’s primary advertising engine. If global ad spending softens or if U.S. President Trump’s trade policies impact the supply chain for high-end semiconductors, Meta could find itself locked into a high-cost structure with diminishing revenue flexibility. For now, the company is betting its balance sheet on the premise that being the biggest spender in AI is the only way to remain the biggest player in social media.
Explore more exclusive insights at nextfin.ai.
