Goldman Sachs: US Data-Center Power Demand to Double by 2027

Deepak Jain — Tue, 19 May 2026 16:00:00 +0000

Goldman Sachs, the US investment bank, has published a projection that electricity demand from US data centers will double by 2027, according to a report circulated on May 19, 2026. The forecast frames the artificial-intelligence computing buildout not as a niche technology story but as one of the largest near-term drivers of US electricity consumption.

Executive Summary

The headline claim is simple and stark: the amount of power consumed by US data centers — the facilities that house the servers behind cloud services and AI models — is projected by Goldman Sachs to double by 2027. A doubling over such a short horizon is extraordinary for electricity demand, a category that in the US grew slowly or stayed flat for most of the two decades before the AI boom.

Why it matters: power, not land or chips, has become the binding constraint on data-center expansion. If a major financial institution’s base case is a doubling within roughly a year and a half of the report’s publication, then utilities, grid operators, regulators, and data-center developers are all planning against a demand curve steeper than anything the sector has seen. Forecasts like this one shape capital allocation — transmission projects, generation buildouts, and multi-year power purchase agreements are being underwritten on the strength of exactly this kind of projection.

Power Is Now the Product

For most of the industry’s history, data-center capacity was measured in square feet; today it is measured in megawatts. The Goldman Sachs projection captures that shift: the constraint on AI infrastructure growth is no longer how fast servers can be manufactured, but how fast electricity can be generated and delivered. AI training and inference clusters draw far more power per rack than traditional enterprise computing, which is why demand can double even if the number of buildings grows much more slowly.

A doubling forecast, if it holds, effectively converts every data-center siting decision into an energy-procurement decision. Markets with available grid interconnection — the formal process of connecting a large load to the transmission system — gain a decisive advantage over markets with cheaper land or better fiber routes. That reorders the competitive map for developers and colocation providers alike.

Who Absorbs the Demand — and Who Profits

Utilities and independent power producers are the most direct beneficiaries of a demand doubling: large, creditworthy, around-the-clock loads are the customers grid operators dream of. Transmission builders, transformer and switchgear manufacturers, and backup-power suppliers sit next in line, since delivering twice the load requires physical equipment that is already supply-constrained industry-wide.

The cost side is less comfortable. Rapid demand growth tends to push up wholesale power prices and interconnection wait times, which raises operating costs for every data-center operator — including those serving ordinary cloud and enterprise workloads rather than AI. Residential and industrial ratepayers in data-center-heavy regions may also bear part of the grid-upgrade cost, a tension that is already a live regulatory debate in several US states.

Reading a Bank Forecast Critically

It is worth being precise about what this is: a projection by an investment bank, not a measurement. Demand forecasts for AI infrastructure have varied widely across analysts, and they are sensitive to assumptions about chip efficiency, model sizes, and how much announced capacity actually gets energized on schedule. Goldman Sachs has a research franchise in this area, but banks also have commercial exposure to the energy and technology sectors they cover, so the appropriate posture is neither dismissal nor uncritical adoption.

The strongest reason to take the direction of the forecast seriously — even if the exact multiple proves off — is that it aligns with observable behavior: hyperscale operators signing long-dated power agreements, utilities revising load forecasts upward, and interconnection queues lengthening. Forecasts can be wrong on timing and still be right about the trend that planners must build for.

Background

US data centers spent two decades as a quiet, efficient corner of the electricity system: demand grew, but efficiency gains in servers and facility design largely kept national consumption in check. The generative-AI boom that began in late 2022 broke that equilibrium. AI clusters concentrate enormous electrical loads in single campuses, and cloud providers and specialized developers have been racing to build capacity, turning power availability into the industry’s defining constraint.

Goldman Sachs is one of several major financial institutions now publishing recurring research on data-center energy demand, reflecting how central the topic has become to utility planning, energy markets, and technology investment. Its projections are widely cited by developers, utilities, and policymakers — which is precisely why the assumptions behind them merit as much attention as the headlines.

Source: US Data Center Power Demand Projected to Double by 2027 – Goldman Sachs, a report published May 19, 2026, projecting a doubling of US data-center electricity demand by 2027.

What They Aren’t Saying

What the Release Doesn’t Say

Baseline and units: the report summary does not state the starting figure — doubling from what base year, and measured in terawatt-hours consumed or gigawatts of peak load?
Methodology: how much of the projection rests on announced projects versus modeled AI adoption, and how does it treat efficiency gains in chips and cooling?
Regional breakdown: national doubling would land very unevenly; the summary gives no view on which grids (for example, established data-center corridors versus emerging markets) absorb the growth.
Supply-side answer: the headline addresses demand only — it does not say whether Goldman Sachs expects generation and transmission to keep pace, or at what price.
Sensitivity: no downside scenario is described — what happens to the projection if AI capital spending slows or announced projects are delayed or cancelled?

Frequently Asked Questions

What did Goldman Sachs actually project?

According to the report published May 19, 2026, Goldman Sachs projects that electricity demand from US data centers will double by 2027. The public summary gives the direction and timeline but not the underlying baseline figures or methodology.

Why is data-center power demand growing so fast?

The main driver is artificial intelligence. Training and running AI models requires dense clusters of specialized chips that draw far more electricity per rack than traditional servers, so total power demand can grow much faster than the number of facilities.

What is a data center, in plain terms?

A data center is a specialized building full of servers — the computers that run websites, cloud services, and AI models. They need large, uninterrupted supplies of electricity and extensive cooling, which is why their growth shows up directly in power-grid statistics.

Is doubling by 2027 a realistic timeline?

It is aggressive but directionally consistent with observable trends: rising utility load forecasts, long interconnection queues, and large power contracts signed by cloud operators. Whether the exact multiple lands on schedule depends on how much announced capacity is actually energized in time.

How does this compare with historical US electricity demand growth?

US electricity demand was roughly flat for much of the two decades before the AI boom, as efficiency gains offset growth. A doubling of an entire load category within a few years is a sharp break from that pattern, which is why the forecast is treated as a macro signal.

Who benefits if the projection proves accurate?

Utilities and power producers gain large, creditworthy, always-on customers. Transmission builders and electrical-equipment manufacturers benefit from the required grid buildout. Data-center operators with secured power positions gain a competitive edge over those still waiting in interconnection queues.

Who bears the costs of a demand doubling?

Data-center operators face higher power prices and longer waits for grid connections. Ratepayers in data-center-heavy regions may shoulder part of the grid-upgrade costs, a burden-sharing question regulators in several states are actively debating.

What is grid interconnection and why does it matter here?

Interconnection is the formal process of connecting a large electricity load or generator to the transmission system. It involves engineering studies and upgrades that can take years, so interconnection availability — not land or fiber — is often the gating factor for new data centers.

Should this forecast be taken at face value?

It deserves serious attention but not uncritical adoption. It is a bank projection, not a measurement; analyst forecasts in this area vary widely and depend on assumptions about chip efficiency and project completion rates. The direction is well supported; the precise multiple is inherently uncertain.

Does the forecast say the grid can actually supply this power?

No. The headline addresses demand only. Whether generation, transmission, and equipment supply chains can keep pace — and at what cost — is exactly the question the summary leaves open, and it is the harder half of the problem.

What does this mean for companies buying cloud or colocation services?

Expect upward pressure on pricing and longer lead times for large capacity commitments, especially in constrained markets. Buyers with multi-year capacity needs benefit from contracting early and asking providers specifically about secured power, not just available space.

What does it mean for investors?

The projection supports the investment case for utilities, grid-equipment makers, and power-secured data-center platforms. The offsetting risk is that AI demand forecasts have a wide error band; capacity built against a projection that slips can pressure returns across the chain.

Why is Goldman Sachs publishing research on data centers?

Goldman Sachs maintains equity and macro research covering the sectors its clients invest in. Data-center power demand now sits at the intersection of technology, utilities, and industrial markets, making it a natural subject for cross-sector bank research.

Could efficiency improvements blunt the demand growth?

Partly. Each chip generation delivers more computing per watt, and cooling efficiency keeps improving. Historically, though, efficiency gains in computing have been outrun by growth in total workload — more efficient AI tends to mean more AI, not less electricity.

Which regions are most affected?

The report summary gives no regional breakdown, but growth is unlikely to be uniform. Established data-center corridors already face grid constraints, which is pushing new projects toward regions with available power — a key detail the forecast leaves unanswered.