Mark RadarMARK RADAR
中 EN
Event File AI AI Data Centers

AI Data Centers Drive Up U.S. Grid Costs, Sparking Debate Over Consumer Burden

1 reports · First detected 2026-07-14 · Last active 2026-07-14

The rapid growth of artificial intelligence has prompted technology giants to build AI data centers at scale, fueling explosive growth in electricity demand and placing unprecedented strain on the power grid. The shift is critical because upgrades to infrastructure such as transmission lines and substations are treated as public costs. Even if technology companies pledge to pay for their own electricity use, the enormous cost of grid upgrades could still be passed on to households and businesses, raising concerns about the fairness of the burden.

A report by PJM's independent market monitor identified surging power demand from AI data centers as the main driver of higher electricity bills. It estimated that the surge could add as much as $23.1 billion in cumulative system costs across the three capacity-auction years from 2025 to 2028. In response to public discontent, the U.S. Federal Energy Regulatory Commission began developing safeguards and reforms in 2026 to prevent the public from being unfairly saddled with grid-upgrade costs.

All Coverage

1 original reports
BLOCKTEMPO.COM 2026-07-13
AI 資料中心讓美國電網多扛 230 億鎂，一般民眾要無辜分攤

The Backstory

The history behind this event
US Bill Would Make Tech Giants Pay AI Data Center Grid Costs2026-06-25 · 1 reports · similarity 0.82

Rapid growth in electricity use by AI data centers could push the cost of new power plants, transmission lines and distribution upgrades onto consumers through higher rates. The US House Committee on Energy and Commerce is therefore advancing the bipartisan H.R. 9340, the Ratepayer Protection Act, which aims to protect households and small and medium-sized businesses by requiring the large electricity users driving demand to bear those costs.

Gabe Evans and Kathy Castor introduced the bill on June 18, 2026, and a House energy subcommittee approved it on June 24. The Energy and Commerce Committee scheduled its review for 5 p.m. on July 20. The bill would apply to a single site with peak demand of at least 100 MW. It sets no fixed amount, instead requiring users to cover the full cost of new generation and transmission and distribution upgrades, as well as provide financial guarantees.

AI and Crypto Drive US Power Demand, Electricity Rates Could Soar 50% by 20302026-05-19 · 1 reports · similarity 0.82

US electricity demand was broadly flat for nearly 20 years, but the expansion of data centers driven by generative AI, alongside cryptocurrency mining, has made energy-intensive computing a new source of strain on the grid. A North Carolina State University team found that data centers’ share of nationwide electricity consumption rose from 1.9% in 2018 to 4.4% in 2023. If the cost of new generation and transmission capacity is passed on to households, it could increase residential power bills while hampering decarbonization efforts.

The study, led by North Carolina State’s Jeremiah Johnson, was published in Environmental Research Letters on May 12, 2026. Its model estimated that average US power-generation costs would rise by 6% to 29% by 2030, reaching as much as 57% in parts of Virginia, while power-sector carbon emissions could increase by up to 28%. Utilities sought rate increases totaling $31 billion in 2025, indicating that the pressure is shifting to household bills.

Mark Radar｜MARK RADAR
Past Issues・by MarkReadFintech