.jpg)
Copyright © 2026 by DCD. All rights reserved. This article first appeared on April 12, 2026 in Data Center Dynamics.
Global electricity consumption for data centers is set to double, reaching approximately 945 TWh in the next four years. Scaling data centers to meet this insatiable appetite for power is expected to cost nearly $7 trillion. But there is a huge reliability concern: half of data centers experienced at least one impactful outage in the past three years.
This data center boom, thirst for power, and its reliability concerns mean a ‘second wind' for construction, representing about $74 billion in spending on data centers in the U.S. in 2024 alone.
Construction firms are under pressure from data center demands for more efficient project execution, faster technology adoption, and a shrinking workforce. The construction labor shortage means that about 300,000 workers are needed to meet project demand, and fast. This is a massive problem for an industry that is historically slow to adopt technology and still heavily relies on manual processes that are hard to scale and often inefficient.
These converging constraints from both grid infrastructure and construction technology reticence in the age of AI require a complete rethink of these sectors’ role in the data center gold rush.
Hyperscalers, such as Meta, Google, Microsoft, xAI, OpenAI, and Oracle, are not waiting for the grid to catch up. As current infrastructure stands, connecting large-scale data centers to the electric grid is dragging out timelines to seven years. This is untenable for hyperscalers that are racing against the clock to get data centers up and running as quickly as possible.
Hyperscalers are resorting to a ‘bring your own power’ approach, moving off the grid to independently power their own data centers. When data centers stay on the grid, these large customers help spread the fixed costs of towers, poles, and wires across more load, which keeps the per-kilowatt price lower for households and small businesses.
When they leave the grid, ratepayers, mainly American families and small businesses, ultimately shoulder inflated energy costs. The knock-on effects of this trend of hyperscalers moving off the grid to power their data centers include higher cost per kilowatt, larger carbon footprints, reduced resiliency, and reliability.
These converging constraints from both grid infrastructure and construction technology reticence in the age of AI require a complete rethink of these sectors’ role in the data center gold rush.
Grids around the world are woefully out of date. The U.S. Department of Energy has warned that the country’s outdated grid is a direct threat to its energy security, potentially increasing blackouts by 100-fold by 2030 and driving up costs for everyone. Most utilities are ready to modernize the grid. However, permitting delays jeopardizes modernization. Findings show that slashing government permitting timelines by just one year could lead to $22bn in returns on invested capital. Put simply, every year of delay in permitting is a rate increase in disguise.
When grid upgrades are stuck in permitting, hyperscalers default to the only timeline they can control: building their own generation, regardless of system-wide consequences. These permitting delays and hyperscalers going off the grid directly impact construction firms who serve electric utilities. Without guaranteed interconnection capacity, utility projects stall in limbo, tying up capital, labor, and critical equipment while creating manufacturing backlogs and uncertainties for critical components like transformers.
All this adversely impacts the financial performance and the ultimate customer of construction firms that serve utility companies. When the customers change from legacy utility customers to new-age hyperscalers, the expectations to adopt technology that drives innovation and efficiency changes dramatically.
All this – data centers approach to ‘bring your own power’ strategies, permitting bottlenecks, and mounting affordability pressure from rate payers and regulators – means that the pressure on power companies and the construction companies serving the utility industry to embrace technology or be sidelined from the next wave of data center investment has never been higher.
In most regulated markets, utilities recover project costs through rates, passing them directly to ratepayers with a built-in profit. Many legacy contract structures fail to reward outcome and performance. Open‑ended time‑and‑materials for contractors' work can turn longer timelines into bigger invoices, creating a built-in incentive to stretch projects rather than deliver results faster. That structure often rewards spending and inefficiency over savings and high performance, resulting in higher utility bills for ratepayers and slowing technology adoption in finance and operations at construction firms that drives efficiency.
In a ‘bring your own power’ scenario, a new model emerges: co-developed data centers, where hyperscalers fund the infrastructure and utilities bring the industry know-how, regulatory alignment, and operational expertise. In this model, utilities, hyperscalers, and construction firms jointly design, finance, and operate the infrastructure that powers datacenters. By treating financial outcomes and high performance as key metrics in contractor evaluation, utilities can reward construction firms that adopt technology to drive efficiency and performance as key partners in this infrastructure development.
Utilities can then embed AI and real-time analytics into the very fabric of project delivery in all their projects; not just co-development projects with hyperscalers. For instance, hyperscalers and utilities can partner with technology-savvy construction companies. These construction companies can adopt AI-driven accounts payable automation that matches invoices to scope, schedule, and delivery.
This keeps high-value equipment and specialty trade subcontractors moving without costly delays. Additionally, predictive forecasting models that flag stress in cash flow and schedule pressure months in advance, while continuous spend analytics surface abnormal cost drift at the package, vendor, or regional level enhance these efficiencies.
The utility and construction companies already have the tools to do this but are missing the will to deploy them at grid scale. The push for power by hyperscalers, the need for affordable power for American families and small businesses might just be the catalyst the utility industry needs to drive technology adoption by its construction partners that will benefit ratepayers and the industry as a whole. Through this approach, projects are more predictable, auditable, and significantly less expensive to build.
For utilities, this means better capital efficiency and regulatory alignment. For hyperscalers, it means faster and more reliable access to clean power. For construction firms and their CFOs, it means a clearer line of sight between project performance, technology adoption, and long-term financial performance. And for ratepayers, it means something even more fundamental: downward pressure to more affordable electric rates instead of surprise hikes every year.
Check out the original article here
Copyright © 2026 by DCD. All rights reserved. This article first appeared on April 12, 2026 in Data Center Dynamics.
Global electricity consumption for data centers is set to double, reaching approximately 945 TWh in the next four years. Scaling data centers to meet this insatiable appetite for power is expected to cost nearly $7 trillion. But there is a huge reliability concern: half of data centers experienced at least one impactful outage in the past three years.
This data center boom, thirst for power, and its reliability concerns mean a ‘second wind' for construction, representing about $74 billion in spending on data centers in the U.S. in 2024 alone.
Construction firms are under pressure from data center demands for more efficient project execution, faster technology adoption, and a shrinking workforce. The construction labor shortage means that about 300,000 workers are needed to meet project demand, and fast. This is a massive problem for an industry that is historically slow to adopt technology and still heavily relies on manual processes that are hard to scale and often inefficient.
These converging constraints from both grid infrastructure and construction technology reticence in the age of AI require a complete rethink of these sectors’ role in the data center gold rush.
Hyperscalers, such as Meta, Google, Microsoft, xAI, OpenAI, and Oracle, are not waiting for the grid to catch up. As current infrastructure stands, connecting large-scale data centers to the electric grid is dragging out timelines to seven years. This is untenable for hyperscalers that are racing against the clock to get data centers up and running as quickly as possible.
Hyperscalers are resorting to a ‘bring your own power’ approach, moving off the grid to independently power their own data centers. When data centers stay on the grid, these large customers help spread the fixed costs of towers, poles, and wires across more load, which keeps the per-kilowatt price lower for households and small businesses.
When they leave the grid, ratepayers, mainly American families and small businesses, ultimately shoulder inflated energy costs. The knock-on effects of this trend of hyperscalers moving off the grid to power their data centers include higher cost per kilowatt, larger carbon footprints, reduced resiliency, and reliability.
These converging constraints from both grid infrastructure and construction technology reticence in the age of AI require a complete rethink of these sectors’ role in the data center gold rush.
Grids around the world are woefully out of date. The U.S. Department of Energy has warned that the country’s outdated grid is a direct threat to its energy security, potentially increasing blackouts by 100-fold by 2030 and driving up costs for everyone. Most utilities are ready to modernize the grid. However, permitting delays jeopardizes modernization. Findings show that slashing government permitting timelines by just one year could lead to $22bn in returns on invested capital. Put simply, every year of delay in permitting is a rate increase in disguise.
When grid upgrades are stuck in permitting, hyperscalers default to the only timeline they can control: building their own generation, regardless of system-wide consequences. These permitting delays and hyperscalers going off the grid directly impact construction firms who serve electric utilities. Without guaranteed interconnection capacity, utility projects stall in limbo, tying up capital, labor, and critical equipment while creating manufacturing backlogs and uncertainties for critical components like transformers.
All this adversely impacts the financial performance and the ultimate customer of construction firms that serve utility companies. When the customers change from legacy utility customers to new-age hyperscalers, the expectations to adopt technology that drives innovation and efficiency changes dramatically.
All this – data centers approach to ‘bring your own power’ strategies, permitting bottlenecks, and mounting affordability pressure from rate payers and regulators – means that the pressure on power companies and the construction companies serving the utility industry to embrace technology or be sidelined from the next wave of data center investment has never been higher.
In most regulated markets, utilities recover project costs through rates, passing them directly to ratepayers with a built-in profit. Many legacy contract structures fail to reward outcome and performance. Open‑ended time‑and‑materials for contractors' work can turn longer timelines into bigger invoices, creating a built-in incentive to stretch projects rather than deliver results faster. That structure often rewards spending and inefficiency over savings and high performance, resulting in higher utility bills for ratepayers and slowing technology adoption in finance and operations at construction firms that drives efficiency.
In a ‘bring your own power’ scenario, a new model emerges: co-developed data centers, where hyperscalers fund the infrastructure and utilities bring the industry know-how, regulatory alignment, and operational expertise. In this model, utilities, hyperscalers, and construction firms jointly design, finance, and operate the infrastructure that powers datacenters. By treating financial outcomes and high performance as key metrics in contractor evaluation, utilities can reward construction firms that adopt technology to drive efficiency and performance as key partners in this infrastructure development.
Utilities can then embed AI and real-time analytics into the very fabric of project delivery in all their projects; not just co-development projects with hyperscalers. For instance, hyperscalers and utilities can partner with technology-savvy construction companies. These construction companies can adopt AI-driven accounts payable automation that matches invoices to scope, schedule, and delivery.
This keeps high-value equipment and specialty trade subcontractors moving without costly delays. Additionally, predictive forecasting models that flag stress in cash flow and schedule pressure months in advance, while continuous spend analytics surface abnormal cost drift at the package, vendor, or regional level enhance these efficiencies.
The utility and construction companies already have the tools to do this but are missing the will to deploy them at grid scale. The push for power by hyperscalers, the need for affordable power for American families and small businesses might just be the catalyst the utility industry needs to drive technology adoption by its construction partners that will benefit ratepayers and the industry as a whole. Through this approach, projects are more predictable, auditable, and significantly less expensive to build.
For utilities, this means better capital efficiency and regulatory alignment. For hyperscalers, it means faster and more reliable access to clean power. For construction firms and their CFOs, it means a clearer line of sight between project performance, technology adoption, and long-term financial performance. And for ratepayers, it means something even more fundamental: downward pressure to more affordable electric rates instead of surprise hikes every year.
Check out the original article here
Hari Vasudevan, PE, is a serial entrepreneur and engineer focused on AI-driven solutions for utilities, construction, and storm response. As Founder and CEO of KYRO AI, he leads the development of AI-powered software that helps utility, vegetation, and field service teams digitize operations, improve storm response and restoration, and reduce operational risk. He also serves as Vice Chair and Strategic Advisor for the Edison Electric Institute’s Transmission Subject Area Committee and holds bachelor’s and master’s degrees in civil engineering with professional engineering licensure in multiple states.
.jpg)
.jpg)
.jpg)