Microsoft and Alphabet Compete Over AI-Driven Growth

SmartInvestor reports Google Cloud revenue grew 63% year on year to US$20 billion in 1QFY2026 and that Google's backlog nearly doubled sequentially to US$462 billion. SmartInvestor reports Alphabet is deploying Gemini across its ecosystem, including Pixel devices and Maps, and that Alphabet has released a Gemini Enterprise agent platform described as enabling customers to "build, orchestrate, govern and optimise agents." The Guardian quotes Alphabet CEO Sundar Pichai saying 2026 is off to a "terrific" start.

SmartInvestor reports Microsoft's Intelligent Cloud revenue was US$34.7 billion in 3QFY2026, up 30% year on year, with Azure and related cloud services growing 40%. SmartInvestor reports Microsoft has launched the Maia 200 AI accelerator, deployed Cobalt server CPUs across regions, and is on track to double its data-center footprint in the next two years. SmartInvestor also reports Microsoft's Productivity and Business Processes segment generated US$35 billion, up 17% year on year, driven in part by Microsoft 365 Copilot adoption; SmartInvestor notes Accenture committed 740,000 Copilot seats.

Coindesk reports four Mag 7 companies (Microsoft, Alphabet, Meta, Amazon) are expected to spend roughly US$650 billion on AI infrastructure in 2026. The Financial Times reports a related aggregate figure of US$725 billion in planned AI spending across Big Tech. Coindesk and the Guardian summarise that quarterly results reinforced investor focus on AI-linked capex and cloud growth during the recent earnings window.

The reported figures illustrate two concurrent dynamics: accelerating top-line growth in cloud revenue tied to AI workloads, and an associated surge in capital expenditure for custom silicon and data-center capacity. Companies increasingly combine model development, silicon, and cloud services to capture more of the AI stack; SmartInvestor's reporting on Maia 200, Cobalt, and Alphabet's TPUs exemplifies that integration trend. For practitioners, these dynamics mean larger hyperscaler investment in model-serving hardware and more on-prem or hybrid silicon options offered to enterprise customers.

Public reporting of enterprise-scale agent platforms and integrated copilots suggests productisation of AI into workflows is moving from experiment to procurement. The Gemini Enterprise phrasing quoted in SmartInvestor and the scale of Copilot seat commitments reported for Microsoft indicate vendors are packaging model-driven capabilities as enterprise features and volume seat deals, rather than solely as research milestones. Observers should expect continued emphasis on governance, orchestration, and manageability features as purchasing criteria.

Reported revenue and backlog growth favor providers that can both supply large-scale compute and embed models into differentiated enterprise products. The two different numbers reported for Big Tech AI capex (Coindesk's US$650 billion and FT's US$725 billion) show variance in aggregate estimates but a consistent signal of multi-hundred-billion-dollar investment across the sector. That degree of capex supports a multiyear market for data-center hardware, networking, and operational tooling.

Key indicators to monitor include hyperscaler unit-level gross margins on cloud AI workloads, silicon availability and performance disclosures (benchmarks for Maia 200, TPUs, Cobalt), enterprise contract structures for agent and copilot products (seat vs outcome-based pricing), and actual capex spend reconciled in future earnings. Also watch Nvidia and other chipmakers' earnings and supply updates, since Coindesk flagged Nvidia's results as the next major test for investor sentiment on AI infrastructure.

Editorial analysis: The narrative above is synthesised from public reporting by SmartInvestor, Coindesk, the Guardian, and the Financial Times. Where sources reported figures or quotes, those items are attributed in the body. Companies' internal intentions, undisclosed roadmaps, or unreported customer negotiations are not inferred here.