The earnings season has returned, and Big Tech companies, including Microsoft, Amazon, and Google, have reported results that point to a single, unmistakable trend, i.e., spending on AI infrastructure continues to accelerate.
Amazon, Alphabet, Microsoft and Meta together poured roughly $120 billion into capital expenditures in the December quarter alone.
The Financial Times reported that this outlay could exceed $660 billion over the full year, surpassing the annual economic output of countries such as the United Arab Emirates, Singapore and Israel. NVIDIA gained nearly 8% on Friday as renewed Big Tech investment reinforced optimism around AI infrastructure.
Google Takes Lead
In the December 2025 quarter, Google and Amazon reported fourth-quarter results, while Microsoft posted its second quarter of the 2026 financial year.
AWS reported revenue of $35.6 billion, up 24% year-on-year. Google Cloud posted sharper growth, rising 48% to $17.75 billion. Microsoft said Azure and other cloud services grew 39% during the same period.
Microsoft’s wider cloud business also expanded strongly. Intelligent Cloud revenue reached $32.9 billion, up 29% year-on-year, while total Microsoft Cloud revenue rose 26% to $51.5 billion.
Among the three, Google Cloud delivered the fastest growth. Its annualised revenue now stands at $70 billion. Backlog surged to $240 billion, up 55% sequentially and more than doubling year over year, signalling that large enterprises are committing to multi-year AI commitments.
Much of this momentum is AI-driven. CEO Sundar Pichai said nearly 75% of Google Cloud customers use its vertically integrated AI stack, spanning custom chips, models, platforms, and enterprise AI agents. AI-led customers consume 1.8 times more products, accelerating revenue expansion and deepening platform lock-in, he added.
Alphabet CFO Anat Ashkenazi said the company expects full-year 2026 CapEx to be $175 billion to $185 billion, with investments ramping throughout the year. “We’re investing in AI compute capacity to support frontier model development by Google DeepMind, ongoing efforts to improve the user experience and drive higher advertiser ROI in Google services,” she said.
Despite Google’s rapid revenue growth, third-party research confirms that AWS still remains the clear market leader by scale. Synergy Research Group estimates that AWS held nearly 29% of the global cloud infrastructure market in Q3 2025, followed by Microsoft Azure at 20% and Google Cloud at 13%.
During the earnings call, Pichai highlighted Google Cloud’s commercial scale, noting that 14 product lines generate over $1 billion in revenue across infrastructure, platforms, and high-margin AI services. He also pointed to Google’s edge in AI infrastructure, citing the industry’s widest compute portfolio ranging from Ironwood TPUs to NVIDIA GPUs.
“Our ten-year track record in building our own accelerators, combined with deep expertise in chips, systems, networking, and software, translates to leading power and performance efficiency for large-scale inference and training,” Pichai said.
More than 120,000 businesses now use Gemini, including Airbus and Honeywell. Gemini Enterprise has crossed 8 million paid seats across 2,800 companies.
Scale Matters For AWS Cloud
AWS may not be the fastest-growing, but it remains the most dominant cloud provider by scale and customer breadth. The company ended the quarter with an annualised revenue run rate of $142 billion, adding $2.6 billion sequentially and nearly $7 billion year-on-year.
Amazon said it expects to invest about $200 billion in capital expenditures across the company, with most directed to AWS to meet surging demand.
Amazon CEO Andrew Jassy said AWS’s custom silicon portfolio, led by Graviton and Trainium, has surpassed a $10 billion annual run rate. Graviton grew by over 50% year-on-year, while the combined portfolio posted triple-digit growth.
AWS continues to win large enterprise and government cloud migrations, Jassy said. Recent customers include OpenAI, Visa, BlackRock, Salesforce, the US Air Force, Adobe, HSBC and the London Stock Exchange Group. “More of the top 500 US startups use AWS as their primary cloud provider than the next two providers combined,” he said.
Graviton processors now deliver up to 40% better price performance than leading x86 chips and are used by more than 90% of AWS’s top 1,000 customers, according to Jassy.
Amazon Bedrock has also gained traction, reaching a multibillion-dollar annualised run rate. Customer spending rose 60 percent quarter on quarter as enterprises scaled generative AI inference workloads.
At AWS re:Invent 2025, the company introduced Frontier Agents, including Kiro autonomous coding agents, AWS DevOps agents and securing agents. Jassy said the tools are already delivering measurable improvements for customers.
Azure’s Enterprise Advantage
Microsoft’s Azure continues to benefit from deep enterprise relationships and strategic AI partnerships. Microsoft CFO Amy Hood said commercial bookings jumped more than 230% in constant currency, driven largely by multi-year Azure commitments from OpenAI and Anthropic.
Microsoft spent $37.5 billion on capital expenditure, with nearly two-thirds directed towards GPUs and CPUs. “Customer demand continues to exceed our supply,” Hood said. The company did not disclose the CapEx for the whole fiscal year.
The company also highlighted its in-house AI chip Maia 200, claiming it delivers three times the FP4 performance of Amazon’s third-generation Trainium and higher FP8 performance than Google’s seventh-generation TPU.
The next phase of the cloud race will depend less on headline growth rates and more on who can turn AI demand into reliable, large-scale deployments. Google has shown it can move fast and win new workloads. Sustaining that lead, however, will require proving that it can deliver consistently at scale quarter after quarter.
Your reaction
Discussion
What's your take on this story?
No comments yet — be the first to share your view.