The Outlook for Data Center Power Demand as AI Token Use Grows

September 1, 2026

AI Summary

5 min read

The Outlook for Data Center Power Demand as AI Token Use Grows

When the last episode on AI and power demand aired, the question was whether the grid could handle what was coming. "That question has now been answered in basically the least comforting way possible," says Alison Nathan: "partly, unevenly and expensively." Since then, forecasts have risen again, and the binding constraints have migrated from megawatts toward turbines, transformers, transmission, and tradespeople—while regulators, ratepayers, and neighbors have taken seats at the table.

The Demand Picture Has Gotten Bigger

Carly Davenport's team recently raised their US power demand forecast to 3.5% compound annual growth through 2030, up from 3.2%. Year to date, actual demand is already running above 4%. The main driver: a jump in data center power demand projections from 83 gigawatts to 108 gigawatts by 2030, based on new projects entering development queues and existing data centers running at higher utilization. Vacancy rates across major US data center markets have fallen to 1–2%, down from a range of 2–7% in recent years.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of Exchanges

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:05) **The Scale of AI-Driven Power Demand** - The episode opens with a striking comparison: AI is adding power demand equivalent to Japan, the world's fifth-largest power consumer.
2 (00:56) **Why Demand Forecasts Keep Rising** - George observes that the demand cycle continues to stretch the supply chain "all the way back to the base elements."
3 (02:56) **U.S. Power Demand Forecast Raised to 3.5% CAGR** - Carly explains the upward revision from 3.2% to 3.5% CAGR through 2030.
4 (04:36) **Efficiency Gains Are Real, But Demand Overwhelms Them** - Brian notes that hyperscaler spending projections have surged from $1.5 trillion to $2.1 trillion for 2029.
5 (07:26) **The Token Efficiency Paradox** - George and Brian discuss how cheaper, more efficient tokens expand the economically addressable market for AI.
6 (08:37) **Regional Hotspots: Where U.S. Demand Is Growing** - Carly maps the current and future data center markets.
7 (10:46) **Why Regulated Utilities Are Winning in MISO** - Carly explains that regulated utilities offer a "one-stop shop" for data center developers, handling everything from power plants to grid connections.

+ Full timestamped outline available in the app

Show Notes

Global data center power demand could rise 170% by 2030 from 2025 levels according to Goldman Sachs Research’s Brian Singer and Carly Davenport. They joined hosts Allison Nathan and George Lee on the Goldman Sachs Exchanges podcast to  discuss how data centers will source that power, the key bottlenecks, and the impact of growing pushback from some local communities.

Key takeaways:

The rise in power demand from AI more broadly is forecast to be the equivalent of adding another Japan to the world’s power consumption by 2030 compared to the start of 2024.
Goldman Sachs Research sees seven key binding constraints for data center power today including price, policy, and people. Delays for connecting data centers to regional US power grids are as long as seven years.
The Mid-Atlantic region should remain the largest US data center power market until at least 2030, but the Midwest is set to overtake Texas for second place.

The opinions and views expressed herein are as of the date of publication, subject to change without notice, and may not necessarily reflect the institutional views of Goldman Sachs or its affiliates. The material provided is intended for informational purposes only, and does not constitute investment advice, a recommendation from any Goldman Sachs entity to take any particular action, or an offer or solicitation to purchase or sell any securities or financial products. This material may contain forward-looking statements. Past performance is not indicative of future results. Neither Goldman Sachs nor any of its affiliates make any representations or warranties, express or implied, as to the accuracy or completeness of the statements or information contained herein and disclaim any liability whatsoever for reliance on such information for any purpose. Each name of a third-party organization mentioned is the property of the company to which it relates, is used here strictly for informational and identification purposes only and is not used to imply any ownership or license rights between any such company and Goldman Sachs.

A transcript is provided for convenience and may differ from the original video or audio content. Goldman Sachs is not responsible for any errors in the transcript. This material should not be copied, distributed, published, or reproduced in whole or in part or disclosed by any recipient to any other person without the express written consent of Goldman Sachs. Disclosures applicable to research with respect to issuers, if any, mentioned herein are available through your Goldman Sachs representative or at http://www.gs.com/research/hedge.html.

Learn more about your ad choices. Visit megaphone.fm/adchoices