Meta Platforms reported fourth-quarter 2025 financial results on January 29, 2026, that exceeded Wall Street expectations while announcing capital expenditure guidance of $115 billion to $135 billion for 2026—nearly double the $72 billion spent in 2025. The massive AI infrastructure investment represents CEO Mark Zuckerberg’s conviction that dominance in artificial intelligence will determine competitive positioning for decades.
Fourth-quarter revenue reached $59.9 billion versus consensus estimates of $58.4 billion, representing 24% year-over-year growth driven by advertising across Facebook, Instagram, WhatsApp, and Threads. Earnings per share of $8.88 surpassed the $8.19 analyst consensus. For first-quarter 2026, Meta guided revenue to $53.5 billion to $56.5 billion, well above the $51.3 billion consensus estimate.
Despite the eye-watering spending projections, Meta shares jumped approximately 10% in after-hours trading as investors accepted Zuckerberg’s argument that current AI infrastructure investment creates durable competitive advantages justifying short-term margin compression. This positive reception marked a shift from October 2025, when Meta’s stock declined after first signaling substantial 2026 spending increases.
The capital expenditures fund construction of massive data centers, procurement of hundreds of thousands of AI accelerator chips primarily from Nvidia, development of energy infrastructure including 6.6 gigawatts of power capacity, and hiring of world-class AI research and engineering talent. Additionally, Meta guided total expenses to $162 billion to $169 billion for 2026, substantially above the $151 billion analyst consensus.
Zuckerberg reorganized Meta’s AI efforts under “Meta Superintelligence Labs,” most dramatically purchasing 49% of Scale AI for $14.3 billion and recruiting Scale CEO Alexandr Wang as Meta’s Chief AI Officer. This represented one of the largest AI-related acquisitions in history and signals willingness to deploy capital aggressively to compete with OpenAI, Google DeepMind, and Anthropic.
The strategic rationale relates to Meta falling behind rivals in 2025 frontier AI model releases. The Llama family of open-weights models was generally viewed as less capable than competing systems, creating both reputational risks and potential business vulnerabilities if competitors develop superior advertising platforms or recommendation systems.
Financial analysts remain divided on the spending trajectory. Bulls argue Meta’s advertising business generates sufficient cash flow to self-fund the AI buildout, that the company has demonstrated ability to monetize AI through improved ad targeting, and that Zuckerberg’s founder-mode governance enables long-term investments insulated from short-term shareholder pressure.
Bears counter that spending increases lack clear evidence of commensurate returns and that Meta lacks the cloud services revenue stream that allows peers like Amazon, Google, and Microsoft to offset AI infrastructure costs by selling capacity to external customers. Operating margin fell to 41% from 48% a year earlier as costs and expenses rose 40% year-over-year.
Meta explicitly stated it expects to deliver operating income above 2025 levels in 2026 despite massive spending, suggesting revenue growth will outpace expense growth sufficiently to maintain year-over-year profit improvement. If delivered, this would demonstrate AI spending coexists with financial discipline rather than representing profitless pursuit of uncertain commercial value.
The announcement coincides with similar spending trajectories from other technology giants. Microsoft is expected to spend approximately $99 billion on capital expenditures in fiscal 2026. Google parent Alphabet is projected to exceed $115 billion. Amazon Web Services has guided to more than $146 billion.
