Citi Lifts Big Tech AI Buildout Forecast by $500B, Now at $2.8T
Citigroup now expects Big Tech to spend more than $2.8 trillion on AI infrastructure through 2029, up from its earlier $2.3 trillion call. Microsoft, Amazon and Alphabet are leading the spending, but Howden data shows US data center power demand has grown 19% a year since 2000 while grid capacity crawled at 2%.
- Original source: cryptobriefing.com.
- The source published it on Nov 22, 2026 at 5:18 PM.
- No second outlet has reported it yet. Treat it as a single-source claim.
Citigroup just raised its AI infrastructure forecast for Big Tech, lifting the estimate by $500 billion to more than $2.8 trillion through 2029, up from a prior call of $2.3 trillion. The bank says the money is arriving sooner than it had modeled, and it covers data centers, GPUs, networking gear and power infrastructure. Microsoft, Amazon and Alphabet are leading the push, with enterprise appetite for AI tools driving the early surge.
Near-term numbers moved too. Citi now sees hyperscaler AI capital expenditures hitting $490 billion by the end of 2026, compared with its earlier estimate of $420 billion. About half the $2.8 trillion, roughly $1.4 trillion, is expected to land in the US.
The Brookings Institution frames it larger still. It puts US spending on data center construction, power infrastructure and computing equipment at $10.3 trillion between 2025 and 2032, an annual average of 3.63% of US GDP, which Brookings says outpaces the railroads and the interstate highway system.
Power is the snag. Insurance broker Howden found US data center electricity demand has grown 19% a year on average since 2000 while grid capacity expanded just 2% annually. Data centers used 0.1% of grid electricity in 2000, a share Howden estimates reached 5% by 2025, and delays on new projects are already stacking up.
Brookings projects 182.8 GW of additional US data center capacity will be running by the end of 2032. Globally, AI compute demand is seen requiring another 55 GW of power capacity by 2030, at an estimated $50 billion per GW. Howden reckons data centers will need to source up to 58% of their power, or 110 GW, from behind-the-meter solutions by 2030.
For GPU, networking and construction suppliers, the revision implies a longer and larger order pipeline. The gap between 19% demand growth and 2% grid growth is the number to watch: without faster grid expansion or alternative power, some of that spending slips further out.
TopicsCitigroup · Microsoft · Amazon · Alphabet · Brookings Institution · Howden
Source: cryptobriefing.com
Written by AI from verified reporting.
More stories
- Tech & AI cryptobriefing.com 2 sources 0 viewsUS intel: China's spy agency ran a two-year hunt for Nvidia chips
- Tech & AI breakingthenews.net 0 viewsOpenAI Opens the Kimono on Pretraining, Eyes Shared Safety Limits
- Tech & AI cbsnews.com Caught by @kylebrennan 1 viewMicrosoft's Smith: AI Set Off a Patching Race, and More Outages May Follow
- Tech & AI cbsnews.com Caught by @kylebrennan 0 viewsBessent Says Nvidia Still Hasn't Answered China Chip Questions

