Key Findings:
- Maine leads the nation in AI-related energy use per business, with each company consuming an average of 8,642 kWh annually, significantly above the national average.
- Virginia, Colorado, West Virginia, andHawaii round out the top five, each reporting over 8,100 kWh per AI-using business per year.
- These energy levels rival the yearly use of thousands of homes or hundreds of EVs, highlighting AI’s growing power demands.
Some states aren’t just using more AI; they’re burning through more power to do it.
Global data center electricity demand is expected to reach 945 billion kilowatt-hours by 2030, according to the IEA.
That's nearly 71 times the total electricity the U.S. consumed in 2025, which was just over 13.3 billion kWh.
AI is expected to be the main driver of this surge, putting added pressure on already strained power infrastructure.
Electricity use from AI-optimised data centres is set to more than quadruple by 2030.
— International Energy Agency (@IEA) May 13, 2025
Government & industry leaders will discuss this key issue – and many more – at our Summit on the Future of Energy Security this week in London 👉 https://t.co/xOPzZnKlLzpic.twitter.com/hdD2JQvKrb
Combined, businesses across all 50 states are using over 348,000 kWh per AI-enabled company on average, revealing the mounting scale of the tech's energy footprint nationwide.
Some states are already seeing AI-related electricity use rise faster than their systems can support.
To pinpoint where the energy burden is heaviest, DesignRush ranked all 50 states by average AI energy use per business.
Top 5 States with the Highest AI Energy Use per Business
-content.jpg "Top 5 States With The Highest AI Energy Use Per Business")
1. Maine
- Average energy use per AI business: 8,642 kWh
Maine ranks first in the nation for electricity use per AI-enabled business, even though it has only around 5,000 such companies.
In total, these firms consume over 43 million kWh each year, which is enough to power more than 4,000 homes.
On average, each business consumes as much electricity as nearly eight household refrigerators running nonstop for a full year.
This level of energy use is significant, especially for a state with modest tech infrastructure.
2. Virginia
- Average energy use per AI business: 8,394 kWh
Virginia is among the most active AI states, with over 42,000 businesses using AI systems.
Together, they consume more than 360 million kWh annually, equivalent to the annual power use of roughly 36,000 homes.
But Virginia’s data center boom has amplified the strain on both its electric grid and water resources.
Digital Dominion: Virginia’s place in the race for the data center future $RKLBhttps://t.co/3NrbgnsnFg
— FranciscoSpace5🚀🔬 (@FranciscoSpace5) June 12, 2025
The state is experiencing “unprecedented energy demand,” with data centers, many supporting AI workloads, becoming major electricity consumers.
The State Corporation Commission has warned that this growth could outpace infrastructure upgrades and affect service reliability.
Additionally, water demands for cooling in AI-driven data centers are raising sustainability concerns, especially in densely packed tech hubs like Loudoun County.
3. Colorado
- Average energy use per AI business: 8,289 kWh
Colorado has the highest AI adoption rate among the top five states at 11%.
Its AI businesses consume more than 350 million kWh annually.
Each company uses enough power to run a medium-sized office building’s HVAC system for a year.
But the state’s growing number of AI-equipped data centers has triggered warnings about electricity shortages and water stress.
Local utilities and policymakers are grappling with how to manage this growth.
Many of these centers require vast electrical capacity and significant amounts of water for cooling, putting added pressure on an already resource-constrained region.
4. West Virginia
- Average energy use per AI business: 8,202 kWh
West Virginia, despite having only about 2,400 AI-using businesses, ranks fourth in per-business electricity use.
These companies collectively use around 19 million kWh per year.
On average, each business consumes as much energy as operating nearly 30 industrial-grade laser printers for an entire year.
This high output per business points to less optimized operations in smaller, rural economies.
5. Hawaii
- Average energy use per AI business: 8,126 kWh
Hawaii’s 3,030 AI-enabled businesses consume more than 24 million kWh of electricity each year.
That is enough to power an average hospital for several months.
Each business uses roughly the same energy as 27 desktop computers running eight hours a day, every day of the year.
Given the state's high utility rates and island grid constraints, this level of consumption may carry steeper financial and environmental costs.
Why These Numbers Matter
Energy costs tied to AI are rising, and the variance across states is more than a technical detail.
Electricity use from data centers could more than double by 2030, with AI as the main driver, according to projections from the International Energy Agency.
A recent analysis by The Verge echoes that outlook, noting that AI’s energy appetite is already rivaling that of energy-intensive sectors like cryptocurrency mining.
AI could consume more power than Bitcoin by the end of 2025 https://t.co/nPcONjeG0c
— The Verge (@verge) May 29, 2025
This sharp increase is changing how energy planners and business leaders think about AI deployment, not just in terms of performance, but cost and sustainability too.
States like Virginia and Colorado are emerging case studies in how AI-related infrastructure can push local utilities to the brink.
This trend raises important questions about how to sustainably support tech expansion without overwhelming essential resources.
Methodology
This analysis quantifies the annual energy footprint of AI use in businesses across all 50 U.S. states.
The goal is to identify where AI-related electricity consumption is most concentrated at the business level.
The following metrics and data sources were used:
- Total Businesses per State
Data on the number of active businesses in each state was sourced from the U.S. Census Bureau (2024). - AI Adoption Rate (%)
AI usage rates were based on business adoption data collected between January and May 2025, using figures from the U.S. Census Business Trends and Outlook Survey (BTOS). - Number of Businesses Using AI
Calculated by applying the AI adoption rate to the total number of businesses in each state. - Employees per Business
The average number of employees per business was obtained from the U.S. Census Bureau to estimate workforce size within AI-using companies. - Total Employees Using AI
Computed by multiplying the number of AI-enabled businesses by the average number of employees per business. - Daily Energy Use per Employee Using AI
According to a May 2025 report by MIT Technology Review, each employee using AI generates approximately 2.9 kilowatt-hours (kWh) of electricity demand per workday. - Annual AI Energy Consumption (kWh)
Assuming 260 working days per year, the annual consumption per state is estimated using the formula:
Employees Using AI × 2.9 kWh × 260 days - Average Energy Use per AI Business
This final metric divides the total annual AI-related energy consumption by the number of AI-using businesses in each state, yielding a per-business energy use estimate.
From surveys to sentiment, turn your data into revenue. Browse leading market research agencies here:
-preview.jpg)
