AWS says most AI cloud capacity booked through 2027, many 2028 slots filled
Amazon Web Services’ chief Andy Jassy told listeners on a video podcast with Ben Horowitz and Erik Torenberg that the company has already booked the bulk of its AI‑focused cloud capacity through 2027. He added that a “good chunk” of the 2028 inventory is also reserved, leaving only a limited pool for on‑demand users and services such as Bedrock and EC2.
Key points
- AWS has booked most AI infrastructure capacity through 2027 and a sizable portion of 2028, according to CEO Andy Jassy
- Amazon plans roughly $220 billion capital expenditure for 2026 and may spend more on data‑center and chip investments
- The forward‑sold model leaves limited on‑demand compute, creating execution risk if construction or chip supply lags
Jassy cited roughly $220 billion of capital expenditure planned for 2026 and hinted the figure could rise as Amazon continues to build data centers and secure advanced chips. The forward‑selling approach, more typical of power‑plant or semiconductor‑fab procurement, pushes execution risk onto AWS: any delay in construction or chip supply could postpone the promised capacity.
The disclosure reshapes the competitive landscape for hyperscale cloud providers, showing that the bottleneck is no longer customer demand but the speed at which Amazon can expand its infrastructure to meet the already‑booked orders.
The story so far
6 episodes →- AWS says most AI cloud capacity booked through 2027, many 2028 slots filledthis story
AWS AI cloud capacity sold out through 2027, 2028 bookings filling fast
vectorwire.ai · 24 October 2026
Loading the full article…
This text was published by vectorwire.ai and written by Vector Wire. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
1source- Reddit discussionreddit.com
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Enterprise & Industry
All →- Google Vault adds filtered exports for Gemini in Workspace · 2 src
- Microsoft Teams adds AI phishing detection for external chats in November · 1 src
- The Agency builds proprietary AI platform for luxury real estate agents · 1 src
- NVIDIA launches Minerva AIOps platform for AI factory observability · 5 src
- Amazon SageMaker adds training session APIs for custom LLM loops · 3 src
Comments
via GitHub Discussions