Three major U.S. tech companies—Microsoft, Meta, and Google—reported rising capital expenditures on AI infrastructure, signaling that their investments are only beginning. Meta now expects to spend between $70 billion and $72 billion this year, up from an earlier forecast of $66 billion to $72 billion. CFO Susan Li indicated spending could be “notably larger” next year. The company’s revenue grew 26 percent year-over-year to $51.24 billion last quarter. CEO Mark Zuckerberg said the focus is on preparing for potential breakthroughs in AI and building capacity ahead of demand.
Meta has also actively recruited AI talent, offering compensation packages worth hundreds of millions of dollars, while cutting roughly 600 positions to optimize its AI teams. The company noted that AI is already benefiting its advertising and virtual reality divisions, though it provided limited specifics.
Alphabet, Google’s parent company, raised its 2025 capital expenditure forecast to $91 billion–$93 billion, up from a prior $75 billion estimate, following a 33 percent revenue increase to $102.3 billion in Q3. Most spending will support AI initiatives and data centers. Google’s cloud business grew 35 percent to $15.15 billion, and its AI app Gemini now has 650 million monthly active users.
Microsoft reported $77 billion in revenue for the quarter ending September 30, up 18 percent, with cloud revenue growing 26 percent. Its capital expenditures reached $34.9 billion, a 74 percent increase from a year ago, mainly for AI infrastructure. CFO Amy Hood said spending will continue to grow in fiscal year 2026, though the company did not give a specific AI-focused forecast. Microsoft’s partnership with OpenAI, including a $13 billion investment, caused a $3.1 billion net income hit this quarter but is expected to drive long-term growth.
Executives emphasized strategic planning for AI infrastructure. Nadella noted Microsoft’s data centers are designed to be flexible, allowing updates to match demand, while the company continues to modernize its systems each year. Analysts note that while these investments increase capacity and efficiency, concerns about a potential AI bubble remain, fueled by large-scale, multi-year projects across the industry.
