Google backs virtual power plant for data center power

Google is funding a virtual power plant through Voltus in PJM to help support data center electricity demand. The deal highlights a growing effort to use grid flexibility, while raising questions about whether households and businesses will participate at scale.

Google has signed a deal with Voltus to help pay for a virtual power plant in PJM, the grid that covers much of the US East Coast. Voltus will group together devices like electric vehicles and smart thermostats, pay customers to participate, and reduce power use or tap stored energy when the grid is stressed. Google will fund the setup, and the added capacity from the project is intended to help run its data centers in the region.

The agreement is one of the clearest examples so far of a tech company using a virtual power plant to help meet rising electricity demand from data centers. Interest in this idea has grown alongside broader discussion of data center flexibility. A study from Duke University found that if data centers agreed to decrease their energy demand for roughly 40 hours per year, about 100 gigawatts’ worth could come online without making new power plants or transmission equipment necessary. The logic is that the grid is built to handle moments of peak demand, so lowering consumption during those periods can create room the rest of the year.

Questions remain about how much flexibility data centers can realistically offer, especially as Artificial Intelligence use expands. Training a model may be delayed or shifted, but customer demand is less flexible, and reducing computing capacity could mean lost revenue. Policymakers are exploring ways to address that. One proposal in the US would allow new data centers to come online years sooner if they agree to lower demand when the grid is nearing its max. A new Texas law also requires large users to switch to backup power or curtail their demand in emergency situations.

Voltus announced in September a program called “Bring your own capacity,” which allows data centers to finance flexibility on their local grid, and Google is the first named customer. Under the new agreement, the company says it will be able to aggregate up to 100 megawatts of distributed energy resources each year. The plant should be operational in 2027, according to Voltus. Google has also made other agreements with utilities across the US to limit or shift its own energy demand, though the company has acknowledged that not every facility can ramp down power use.

The bigger uncertainty may be whether customers will participate in enough numbers. A recent California study on managed electric-vehicle charging found that with no economic incentive, only 1% of EV owners enrolled in managed charging. At ? per month (about 15% of their power bill), only 4.6% did. The companies have not disclosed what they plan to pay participants in this project. That matters because even when money is offered, people may still resist giving up control over electricity use, particularly as about 70% of Americans oppose Artificial Intelligence data centers in their area, according to recent Gallup polling.

63

Impact Score

Artificial Intelligence Forge targets national security research gaps

DARPA and the National Science Foundation are launching Artificial Intelligence Forge to push research on national security problems that commercial development often overlooks. The effort focuses on reliability, interpretability, control, and resilience in high-stakes and contested environments.

SoftBank backs France Artificial Intelligence infrastructure expansion

SoftBank plans a major buildout of Artificial Intelligence infrastructure in France, centered on new data center capacity in Hauts-de-France and industrial partnerships in Dunkirk. The investment underscores France’s push to become a leading European hub for high-performance compute.

Nvidia pushes Artificial Intelligence PCs into the mainstream

Nvidia is positioning Artificial Intelligence PCs as a new phase for personal computing, with Jensen Huang arguing that the category could reshape how laptops and desktops handle demanding software. The shift centers on running more Artificial Intelligence features directly on devices instead of depending heavily on the cloud.

Nvidia unveils RTX Spark Windows PCs

Nvidia introduced RTX Spark, its first fully integrated chip for Windows laptops and desktops, aiming to bring local Artificial Intelligence agents, gaming, and productivity workloads onto its own hardware stack. The new systems are expected this fall from major PC makers, with premium pricing and no detailed benchmarks yet disclosed.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.