One Year In, Trump's Data Center Order Is Costing Communities Dearly

When President Trump signed Executive Order 14318 on July 23, 2025, the White House promised a "golden age for American manufacturing and technological dominance." One year later, the record shows something different: a policy that systematically strips environmental review from some of the largest industrial projects in American history, opens public lands and even Superfund sites to private AI companies, and leaves ordinary ratepayers holding the bill.

What the order actually does

The order — which revoked the prior administration's AI infrastructure directive and replaced it with a far more aggressive framework — applies to data centers drawing more than 100 megawatts of new load or backed by at least $500 million in capital spending, along with the pipelines, transmission lines, gas turbines, and even coal power equipment that serve them.

For these "Qualifying Projects," the order directs agencies to create new categorical exclusions under the National Environmental Policy Act — the mechanism that lets projects skip environmental review entirely. It declares that federal subsidies covering less than half of a project's cost are presumed not to trigger NEPA at all. It instructs the EPA to modify regulations under the Clean Air Act, Clean Water Act, and Superfund law where they "impact the development" of data centers. And it orders the EPA to identify Superfund and Brownfield sites — the nation's inventory of contaminated land — for data center reuse, guidance the agency delivered in January 2026.

Read the definitions section closely and another detail jumps out: "covered components" explicitly include natural gas turbines and coal power equipment. This is not a clean-energy buildout. It is a permitting fast lane for fossil-fueled computing, as legal analysts noted when the order was signed.

Ratepayers are already paying

The clearest cost so far shows up on electric bills. In PJM, the grid region serving 65 million people from Illinois to Virginia, capacity auction prices rose roughly tenfold in recent auctions, hitting federal price caps two years running. PJM's independent market monitor attributed about 63% of one year's increase to data center demand — roughly $9.3 billion recovered from customers in higher rates. E&E News reported data centers drove a 76% surge in PJM power prices.

Households feel this directly. Washington, D.C.-area Pepco customers saw bills jump by an average of about $21 a month starting in June 2025, and consumer advocates project increases approaching $70 a month by 2028 in parts of PJM territory if nothing changes. Consumer Reports has documented the pattern nationally: residential rates climbing while the data centers driving new demand negotiate discounted industrial rates.

Honesty requires a caveat the administration never offers: the picture is contested. An E3 analysis found little evidence of a historical cost shift to residential customers, and PolitiFact rated some Democratic claims about past price impacts as overstated. But the forward-looking research is grim: academic modeling projects wholesale prices rising 6% to 29% nationally by 2030 — up to 57% in the hardest-hit regions — as AI demand compounds. The question is no longer whether the buildout raises costs, but how much and who pays. Executive Order 14318 answers the second question by default: not the trillion-dollar companies building the facilities.

Communities cut out of the process

The order's deepest damage may be procedural. NEPA review is often the only forum where a community learns what is being built next door — how much water it will draw, how loud the turbines will run, what happens to local air quality. Categorical exclusions and the 50% financial-assistance presumption are designed to close that forum.

The consequences are surfacing in court. The administration's approval of a data center on public land — reportedly without a project-specific environmental assessment and without resident input — now faces a lawsuit. Advocacy groups like Food & Water Watch argue the administration's approach amounts to overriding local objections on behalf of developers. State and local governments are pushing back with moratoriums and zoning fights across the country.

And the backlash is not a partisan phenomenon. Polling this summer shows majorities would oppose a data center in their own community, with concerns about energy costs, water use, and environmental impacts outweighing promised economic benefits. Roughly 100 projects were cancelled or stalled in the past year as local opposition quadrupled. Even Rep. Marjorie Taylor Greene — hardly a critic of this White House — broke with the president over the AI buildout, citing state rights, jobs, the environment, and "critical water supply."

When Pennsylvania farm towns, D.C. ratepayers, environmental lawyers, and MAGA congresswomen all land in the same place, that is not politics. That is a policy failing on its own terms.

The corruption question

None of this happens in a vacuum. The companies benefiting from expedited permits, federal land leases, and the financial support the order instructs Commerce to arrange — loans, grants, tax incentives, offtake agreements — are the same firms whose executives have cultivated this administration assiduously. The order creates a discretionary designation power: any project the Secretaries of Defense, Interior, Commerce, or Energy chooses to bless becomes a "Qualifying Project" entitled to the fast lane. Discretion of that breadth, exercised on behalf of the wealthiest corporations on earth, with environmental review stripped away and the public locked out, is precisely the environment in which favoritism thrives — whatever one thinks of AI itself.

A serious industrial strategy would pair the buildout with binding ratepayer protections, enforceable water commitments, and community consent. This order does the opposite: it socializes the costs and privatizes the discretion. One year in, the returns are visible on your electric bill.

This is an opinion piece. The factual claims above are drawn from the cited government documents, market analyses, and news reporting.