Data centers and crypto could hike power costs by 57% by 2030

2026-05-20 · Source: Dataconomy · Field: Energy & Utilities — Energy Markets & Policy, Cloud Computing & IT Infrastructure, Environmental Science & Earth Systems · Depth: Novice, quick

Summary

New research from North Carolina State University, Carnegie Mellon University, the University of Pittsburgh, and the University of Toronto predicts significant increases in U.S. electricity costs and CO2 emissions by 2030. The study, published in Environmental Research Letters, forecasts that electricity demand from data centers and cryptocurrency mining could raise power costs in some regions by up to 57%, with a national average increase between 6% and 29%. This expansion is also projected to increase CO2 emissions by as much as 28% by 2030 compared to a scenario without data center growth. Researchers used computational modeling to analyze hourly energy supply and demand across 26 contiguous U.S. regions, identifying data center expansion as the primary driver. Price hikes are expected to be most severe in states like Virginia, Pennsylvania, and New York, with geographical concentration of data centers exacerbating the issue.

Key takeaway

For policymakers and utility executives planning future energy infrastructure, this research indicates your current strategies may be insufficient to mitigate rising electricity costs and CO2 emissions. You should urgently reassess near-term power generation plans, considering the significant impact of concentrated data center expansion. Ignoring these projections risks undermining decades of carbon reduction efforts and imposing substantial cost burdens on consumers by 2030.

Key insights

Data center and cryptocurrency expansion threatens to reverse U.S. power sector gains, driving up costs and CO2 emissions by 2030.

Principles

• Increased power demand from data centers correlates with higher electricity costs and CO2 emissions.
• Geographical concentration of data center expansion intensifies regional price hikes.
• Gains in power sector carbon emission reduction are vulnerable to increased demand.

Method

Researchers utilized computational modeling to assess future power demand from data centers and cryptocurrency through 2030, analyzing energy supply and demand hourly for 26 U.S. regions.

In practice

• Regulators and utilities should reassess near-term power generation strategies.
• Government officials must consider data center construction implications.

Topics

• Data Centers
• Cryptocurrency Mining
• Electricity Costs
• CO2 Emissions
• Energy Policy
• Power Grid Modeling

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Policy Maker, Consultant, Research Scientist

Related on AIssential

• Greenhouse gases from data center boom could outpace entire nations — The AI boom's data center expansion is driving a significant, under-recognized surge in natural gas-powered, behind-the-meter emissions.
• Senators want US energy information agency to monitor data center electricity usage — Mandatory data center electricity disclosure is critical for grid planning and consumer cost protection.
• Assessing the Carbon Emissions and Energy Consumption of U.S. Hyperscale Data Centers — US hyperscale data centers disproportionately rely on carbon-intensive electricity, driving significant emissions and energy demand.
• More than 100 UK datacentres plan to burn gas to generate electricity — UK datacentres are increasingly turning to gas power due to grid connection delays, jeopardizing climate targets.
• ‘We should not have to sacrifice’: New York could become first state to temporarily ban large datacenters — New York's proposed datacenter moratorium reflects a growing public and legislative pushback against the unbridled expansion of energy-intensive AI infrastructure.
• Officials hugely underestimated impact of AI datacentres on UK carbon emissions - The Guardian — UK government vastly underestimated AI datacentre carbon emissions, raising climate concerns.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Dataconomy.