Big Tech Reports $650 Billion AI Capex and Cloud Growth

Per The Next Web, Google Cloud grew 63% year on year to $20.02 billion in Q1 2026, beating analyst estimates, and Alphabet reported net income of $62.57 billion, up 81%, with an apparent $36.9 billion unrealised gain on equity securities (The Next Web). The Next Web reports Alphabet recorded $35.7 billion of capital expenditure for the quarter and raised its full-year capex guidance to $180-$190 billion from $175-$185 billion (The Next Web).

The Next Web reports AWS revenue grew 28% to $37.59 billion, its fastest pace in years, with Q1 earnings-per-share of $2.78 versus a $1.62 estimate; The Next Web also notes management has committed roughly $200 billion of capex for 2026 and that trailing twelve-month free cash flow compressed to about $1.2 billion (The Next Web).

Meta: The Next Web and SiliconRepublic report Meta raised its 2026 capex guidance to $125-$145 billion from a prior range, and that shares fell roughly 6% after hours following the update (The Next Web; SiliconRepublic). SiliconRepublic and other outlets note the increase reflects larger data centre and component spending tied to AI capacity expansion (SiliconRepublic).

The Next Web reports combined 2026 AI-related spending across five hyperscalers is now on track to exceed $650 billion, reflecting a broad industry ramp in infrastructure investment (The Next Web).

Industry observers: Companies scaling foundation-model workloads typically shift capex toward racks, networking and power, which raises near-term capital intensity even when top-line revenue grows. Observers such as Forrester quoted by SiliconRepublic highlight persistent questions about the sustainability and utilization rates of large data-centre buildouts (SiliconRepublic). For practitioners, that pattern translates into tighter on-premise procurement cycles, longer lead times for specialised hardware, and renewed emphasis on capacity planning and cost-per-token metrics.

The combination of high revenue growth in cloud businesses and sharply higher capex guidance underlines a contrast analysts and reporters are tracking: revenue gains from AI-driven products are material today, yet the infrastructure bill to support larger generative workloads is also rising materially. The Next Web and SiliconRepublic both frame the quarter as evidence that demand is strong but that hyperscalers face near-term compute constraints-The Next Web quotes Alphabet CEO Sundar Pichai saying the company is "compute constrained in the near term" (The Next Web; SiliconRepublic).

Industry watchers should track these indicators in coming quarters: utilization and backlog metrics for cloud AI services (The Next Web reported Google Cloud backlog above $460 billion), capex-to-revenue ratios, gross margins on AI services as hardware depreciation and energy costs scale, and any company disclosures about queueing, throttling or pricing changes for high-cost inference workloads. Observers will also watch third-party supply signals such as GPU/TPU availability and component cost trends reported by outlets covering the sector.

For practitioners: the reports collectively mean cloud providers are expanding capacity aggressively but still cite capacity bottlenecks; teams planning production-level generative AI deployments should treat lead times and incremental unit costs as first-order risks. Industry reporting across The Next Web and SiliconRepublic provides the current market signals: strong cloud demand, rising capex, and continued investor scrutiny over capital intensity.