> ## Content Index
> Fetch the complete content index at: https://www.theamericanquorum.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# AWS Capacity Crunch Extends Through 2027, Jassy Says
- URL: https://www.theamericanquorum.com/aws-capacity-crunch-extends-through-2027-jassy-says/
- Published: 2026-10-24T06:23:26.000Z
- Updated: 2026-10-24T06:23:26.000Z
- Description: Amazon says most AWS capacity planned through 2027 is already committed, showing how AI demand is outrunning data-center construction while exposing the company to enormous spending and power constraints.
- Author: News Desk
- Tags: Tech

Amazon Web Services has committed most of the cloud-computing capacity it expects to bring online through 2027, with a significant share of 2028 capacity already reserved, Chief Executive Andy Jassy said Friday. The disclosure, first reported by [Reuters](https://www.reuters.com/business/retail-consumer/amazon-cloud-capacity-sold-out-through-2027-ceo-jassy-says-2026-10-23/?ref=theamericanquorum.com), is the clearest indication yet that demand for artificial-intelligence infrastructure continues to outrun the pace at which Amazon can build data centers and connect them to power.

Jassy made the comments in an interview with Andreessen Horowitz co-founder Ben Horowitz and partner Erik Torenberg. In the [podcast transcript](https://podscripts.co/podcasts/a16z-podcast/amazon-ceo-andy-jassy-on-why-amazon-is-investing-220-billion-in-ai?ref=theamericanquorum.com), he described AWS as still leaving meaningful demand unserved even as Amazon plans roughly $220 billion in capital spending this year. The capacity claim refers principally to large blocks reserved by major customers, not a literal shutdown of ordinary AWS services or an inability for every customer to buy computing resources.

## AI demand is changing how cloud capacity is sold

Traditional cloud services were designed around shared pools of computing power that customers could expand or reduce on demand. Training and operating large AI models requires a different scale: dense clusters of advanced processors, high-speed networking, cooling and reliable electricity, often committed under multiyear agreements. A large customer may reserve an entire block before a facility is finished because waiting risks losing access to scarce power and chips.

That distinction explains why AWS can be “largely sold out” years in advance while continuing to add customers and launch services. Amazon is matching future construction to contracted or strongly forecast demand, giving it more confidence that new facilities will generate revenue. Customers gain priority access, but may also accept longer commitments and less flexibility than cloud computing historically promised.

## Amazon is spending heavily before revenue arrives

The buildout requires Amazon to spend long before it can bill customers. Its latest [SEC filing](https://www.sec.gov/Archives/edgar/data/1018724/000101872426000026/amzn-20260630.htm?ref=theamericanquorum.com) shows cash capital expenditures of $53.1 billion in the second quarter and $96.3 billion in the first six months of 2026, primarily for technology infrastructure supporting AWS and for fulfillment capacity. The same filing shows Amazon raised $82.4 billion through short- and long-term debt during that six-month period, while also funding large strategic investments.

Amazon raised its full-year capital-spending plan to about $220 billion after previously projecting $200 billion. The [Associated Press](https://apnews.com/article/b4ce02b4666a35b8975823c5c22072ee?ref=theamericanquorum.com) reported in July that the increase also covers robots, semiconductors and satellites, so not every dollar is destined for AWS. Still, Amazon says most of its technology-infrastructure spending supports cloud growth, and the timing creates an unavoidable gap between cash outlays and revenue.

## Rapid AWS growth supports the investment case

The demand claim is backed by recent operating results. AWS second-quarter sales rose 37% from a year earlier to $42.2 billion, while segment operating income increased to $16.6 billion from $10.2 billion, according to Amazon’s [earnings release](https://ir.aboutamazon.com/news-release/news-release-details/2026/Amazon-com-Announces-Second-Quarter-Results/?ref=theamericanquorum.com). Jassy said AWS had reached a $169 billion annualized revenue run rate, while Amazon’s AI and custom-chip businesses each exceeded a $25 billion run rate.

Those figures show real revenue growth, but they do not guarantee that every planned data center will earn an adequate return. Customers can change their deployment plans, chip efficiency can improve, and lower-cost models can reduce the computing needed for a given task. Long reservations lower demand risk; they do not eliminate construction, financing, technology or execution risk.

## Power is becoming as important as processors

Cloud companies can order servers more quickly than utilities can build generation and transmission. Data centers require firm power, backup systems and grid connections that may take years to permit and construct. Amazon must coordinate land, buildings, cooling, chips and networking so that costly equipment does not sit idle while a facility waits for electricity. That sequence makes power availability a central constraint on how quickly reserved capacity can become billable service.

The national scale is substantial. Lawrence Berkeley National Laboratory estimates that data centers could consume 11.8% of U.S. electricity by 2030, compared with 4.4% in 2023\. Its [updated study](https://bies.lbl.gov/publications/united-states-data-center-energy-2025?ref=theamericanquorum.com) uses equipment-shipment data, device energy models and cooling simulations, but the estimate remains a projection whose outcome depends on deployment speed and efficiency. The report’s scenarios underscore why securing power, rather than purchasing processors alone, increasingly determines how quickly cloud capacity can expand.

## Scarcity gives AWS leverage and creates exposure

A reserved pipeline can strengthen Amazon’s position with suppliers, utilities and customers. It can also support pricing by limiting idle infrastructure. For companies building AI products, however, a multiyear shortage may raise costs, lengthen deployment schedules and encourage the use of several cloud providers or private data centers rather than dependence on one platform.

The key uncertainty is whether today’s commitments translate into durable use. Amazon’s disclosure establishes that major customers are reserving capacity faster than AWS can add it, but it does not reveal contract terms, cancellation protections or the share devoted specifically to AI. Amazon is scheduled to report third-quarter results on Oct. 29, when investors will look for updated AWS growth, spending and cash-flow figures. Those results will help show whether the capacity crunch is converting into revenue quickly enough to justify the scale of the buildout.